跳到正文
研踪PaperTrail

论文库

检索计算机领域论文，按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+2已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-15 00:32 · 全站最近入库：2026-11-14 15:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-14 01:08 — 2026-11-15 01:08（北京时间） · 点击数字查看对应简报；AI 相关包含在新增论文内。

最近完成检查：2026-11-15 00:32最近发现新增版本：2026-11-14 15:09库内论文最新更新时间：2026-11-13 02:59

每分钟刷新已保存的数据。空检查不代表采集异常；新收录可能包含历史论文。AI 导读按新增版本计数。

23,348 已收录论文19,580 AI 相关6,729 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →

找到 56 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-15 00:30（北京时间）
新论文 · v1arXiv:2611.16466

How Should Planning Agents ReSHAPE Their Search Space?

Yilin Zhang, Xinran Zhao, Tong Chen, Ziming Yuan, Tongshuang Wu

原文摘要

Planning agents and benchmarks treat a user's initial specification as fixed. Yet the requests users write exclude plans they would gladly take, since they rarely discover the possible trade-offs they would accept until seeing the options. Agents already revise requests through interaction, but whether this helps planning is unknown. We formalize this step as the ReSHAPE task, which judges a revision by whether its search space matches the acceptable plans and a planner finds one there. Using ReSHAPE Planter, which builds instances from planning benchmarks with a human-validated simulated user, we analyze 5,800 agent-revised specifications in travel and shopping. We find that reshaping matters: one correct trade-off makes the same planner reach an acceptable plan up to 4.6x as often, and search space quality predicts what the planner finds. Current agents, however, admit at most 32% of the acceptable plans, where that trade-off alone admits over 52%, because they ask about preferences instead of showing plans; in travel, showing plans raises F1 by 9.7 points over asking, with two model families. Planning agents should thus treat a request as a hypothesis to test against the user's reactions to concrete plans to recognize user trade-offs.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.16409

Experience-Handoff Reinforcement Learning for Recursive Self-Improvement

Doyoung Kim, Byeongguk Jeon, Kailin Zhang, Sherry Yang

原文摘要

It is common to train Large Language Models (LLMs) for agentic tasks with Reinforcement Learning (RL) from scalar rewards. However, this throws away the fine-grained information about the environment that the agent experienced during the trial (e.g., error logs). To utilize this rich information, recent works wrap the agent in a harness that refines its own inputs through reflection or compaction and hands past experience to future rollouts, but the construction of the handed-off artifacts (e.g., a playbook) is itself not learned. We want the best of both worlds: gradient updates from a scalar reward, while also utilizing environment information to hand off experience. As a concrete testbed, we fix the Agentic Context Engineering (\ace{}) harness and train one model to act, reflect, and curate task-local playbooks. Our method, \acerl{}, evaluates candidate experience-handoff artifacts by their downstream effect on future actor rollouts, and co-optimizes the role-specific behaviors (action generator, insight reflector, and playbook curator) under a fixed scaffold. Across agentic benchmarks, \acerl{} achieves higher held-out performance than baseline methods. The results suggest that agent training should optimize not only the actions in the current rollout, but also the model-written experience that future rollouts inherit.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.35036

Persona Following Is Not Selective Control: The Neutrality Gap in LLM User Simulation

Jiashen Ren, Wenlin Zhang, Bohan Zhang, Xiaopeng Li, Zichuan Fu, Wanyu Wang, Junyi Li, Xiangyu Zhao

原文摘要

Persona prompting is widely used in LLM user simulation, yet it relies on an untested assumption: specifying a target user attribute should shift target behavior while leaving other fixed attributes unchanged. Across eight black-box and five open-weight models, however, changing a target attribute alone also shifts unspecified non-target choices, such as risk preference shifting color choice. We test whether explicitly specifying non-target attributes restores selective control across 477 held-out synthetic worlds and 200 independent items under unspecified, directional, and neutral states. Assigning a clear non-target direction nearly eliminates cross-attribute shifts. Explicitly declaring no preference, however, still leaves 37.7%-76.3% of held-out worlds and 51%-81% of independent non-target items shifting with the target attribute while preserving target efficacy. We call this disparity the neutrality gap. Because directional tests can be passed simply by following a stated preference, the neutral state reveals selective-control failures they miss: following a stated persona does not imply keeping unrelated choices invariant. In simulated investment decisions, incidental brochure color labels shift estimated target risk treatment effects by 25.5 percentage points, even when models accurately report having no color preference. Counterfactual attention interventions show that unwanted influence enters primarily at query time, when option evaluation reads the target persona directly. Blocking query-time attention cuts non-target drift by 72.1%-87.8%, whereas directional declarations suppress drift primarily by overwriting option values rather than rerouting attention. Because target and non-target choices share query-time access, blocking it weakens target efficacy while value overwrite imposes an artificial direction, leaving task-dependent selective access as a central challenge.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.15431

Subadditivity and the failure of strong subadditivity in the Nichols--Wootters family of entropies

Lin Zhang, Wenlin Zhang, Haobo Ma

原文摘要

We prove subadditivity for the Nichols--Wootters family of entropies interpolating between the von Neumann entropy and the subentropy, with all spectra evaluated in one common ambient dimension. Beyond the von Neumann entropy, equality holds exactly when at least one marginal is pure. This settles the subadditivity conjecture concerning subentropy. Every member except the von Neumann entropy fails strong subadditivity, already on a maximally entangled two-qubit state tensored with a mixed qubit. Its strong-subadditivity defect takes both signs on quantum Markov states. For a pure bipartite state tensored with an ancillary state, we identify the defect with a separately concave product gap and determine its exact maximum under rank constraints. For arbitrary bipartite states, we obtain an explicit lower bound in the marginal purities. For two-qubit states, we determine the optimal coefficient multiplying the product of the smaller marginal eigenvalues in every common ambient dimension, including the subentropy coefficient $1/3$. A positive representation by Haar averages of compressed matrix entropies gives the pinching inequality underlying these results and also yields entropy-gain bounds for unital quantum channels.As an application, each entropy equals the classical communication capacity retained after a specified isotropic partial projective measurement, at a prescribed average input state, even when entangled block encodings are allowed.

quant-ph · cs.IT · math-ph

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.39941

插入的块交换下降：交换投影与最小掩码

Block-exchange descent for insertion: commuting projections and minimum masks

Haobo Ma, Wenlin Zhang, Charles E. Hughes

AI 速读 基于摘要

研究shuffle表示中最小插入度在块交换下的下降规律，并给出交换投影下的精确刻画与若干计算机辅助验证结果。

cs.FL · math.CO

arXiv ↗PDF ↗
查看原文摘要

For fixed nonempty words $u,v$, the minimum insertion degree of a shuffle is the least number of inserted-source runs among its representations. We prove that, after a block exchange from a minimum mask of degree $r$, the minimum over masks with commuting source projections is exactly $r-1$; a larger decrease occurs precisely when a noncommuting mask has fewer runs. If either source is unary, every such exchange is exact. We prove an explicit repair for an infinite periodic degree-three family. Computer-assisted results establish first-mask descent through combined length seventeen over every alphabet, degree-three first-mask descent through length thirty, and a uniform repair of all degree-three failures through length thirty, including all further letter identifications. Counterexamples to arbitrary, first, and last minimum-mask choices explain the need for alternative masks. Finally, we prove that minimum iteration depths have no gaps for arbitrary languages. General block-exchange descent and Singleton No-Gap remain open.

新论文 · v1arXiv:2611.13289

MRFoundry: Knowledge-guided agent loops for end-to-end MRI research

Ke Wang, Atakan Topcu, Hossein Momeni, Molin Zhang

原文摘要

AI agents can write code, use scientific software, and revise computational workflows, opening new possibilities for MRI research. Effective assistance requires connecting acquisition physics, data conventions, specialized tools, and scientific validation. We present MRFoundry, an open-source, extensible package that connects MRI knowledge to agent actions and output checks. Project-local memory preserves decisions and evidence, while reviewed contributions allow researchers to improve shared guidance. Researcher-supervised examples demonstrate skill-guided sequence auditing and acquired-data reconstruction, connecting tool use, output checks, and documented decisions. Alongside MRFoundry, we introduce MRI Agent Bench to compare agents with and without the package across models and MRI task domains. Across 240 paired Core evaluations spanning 24 task types and ten models, all-checks-passed completion increased from 67.5% without MRFoundry to 78.3% with it. Across 134 Hard pairs, completion increased from 36.6% to 44.8%. MRFoundry makes MRI workflow expertise available as reusable guidance that researchers can inspect, adapt, and extend within existing agent and scientific software environments. MRFoundry is openly available at https://github.com/KeWang0622/MRFoundry.

eess.IV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2606.15874

LLM-as-Code: Agentic Programming for Agent Harness

Junjia Qi, Zichuan Fu, Jingtong Gao, Wenlin Zhang, Hanyu Yan, Xian Wu, Xiangyu Zhao

原文摘要

Every major LLM agent framework gives the LLM the role of orchestrator; the model decides what to do next, when to call tools, and when to stop. We argue that token explosion, control-flow hallucination, and unreliable completion are not implementation bugs but architectural consequences of assigning the deterministic work of looping, branching, and sequencing to a probabilistic system. A better prompt or a stronger model cannot guarantee the reliability of the LLM agent. We therefore propose Agentic Programming, in which the program governs all control flow, and the LLM is itself part of it, an adaptive component we call LLM-as-Code and invoke only where a task calls for reasoning or generation. Within each call the model keeps full flexibility, but it cannot alter the program's execution path. With control in the program, the LLM's context is built from the execution history's call tree and forms a directed acyclic graph (DAG). Each call's context length is then determined by its call depth rather than by accumulation over steps. A case study of computer-use agents shows that the design is practical, not just a theoretical stance, substantially improving the stability of long visual operation sequences.Code is available at https://github.com/FuzzyKuji/OpenProgram.

cs.AI · cs.SE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12383

PCTac: Mechanochromic Photonic-Crystal Tactile Sensor for Dexterous Manipulation

Ruilin Zhang, Xuanshuo Liu, Junzhe Dai, Kefei Wu, Xuhao Qin, Ziyuan Jiao, Wanlin Li, Yurui Qu 等 9 位作者

原文摘要

Modern visuotactile sensors primarily sense contact by imaging the deformation of a compliant interface. Using techniques such as photometric stereo, they can recover detailed contact geometry, often represented as surface normals or depth. However, force-related information typically still needs to be inferred indirectly through calibration, geometric reconstruction, or learned mappings. Mechanochromic photonic-crystal (PC) materials offer an alternative physical encoding mechanism: contact loading changes their periodic microstructure and spectral response, directly transforming local stress into a spatially resolved structural-color field. Despite this potential, PC materials have rarely been explored as visuotactile sensors for closed-loop robot learning. We present PCTac, an open-source mechanochromic PC fingertip that uses a dense hue field as its tactile observation. Loading characterization demonstrates a repeatable relationship between the applied normal load and the observed hue response. We further integrate this representation directly into a Diffusion Policy and evaluate it on three real-robot manipulation tasks. Across 90 physical rollouts, PCTac Hue achieves 68 successes (75.6\%), compared with 60 (66.7\%) for PCTac RGB, 57 (63.3\%) for Wedge RGB, and 29 (32.2\%) for vision-only. PCTac Hue also achieves the highest observed success rate on each evaluated task, including Perturbed Test-Tube Insertion, where it achieves 16/30 successes versus 13/30 for tactile RGB baselines and 1/30 for vision-only. These results demonstrate the potential of mechanochromic structural color as a compact, load-sensitive tactile representation for robot learning and closed-loop manipulation. Project page: https://bey0ndt1me.github.io/PCTac-page/.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2510.11370

Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers

Wenhan Ma, Hailin Zhang, Liang Zhao, Yifan Song, Yudong Wang, Zhifang Sui, Fuli Luo

原文摘要

Reinforcement learning (RL) has emerged as a crucial approach for enhancing the capabilities of large language models. However, in Mixture-of-Experts (MoE) models, the routing mechanism often introduces instability, even leading to catastrophic RL training collapse. We analyze the training-inference consistency of MoE models and identify a notable discrepancy in routing behaviors between the two phases. Moreover, even under identical conditions, the routing framework can yield divergent expert selections across repeated forward passes. To address this foundational inconsistency, we propose Rollout Routing Replay (R3), a method that records routing distributions from the inference engine and replays them during training. R3 significantly reduces training-inference policy KL divergence and mitigates extreme discrepancies without compromising training speed. Extensive experiments on various settings confirm that R3 succeeds in stabilizing RL training, preventing collapse and outperforming methods such as GSPO and TIS. We believe this work can offer a new solution for stabilizing RL in MoE models.

cs.CL · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.11929

Can LLM Agents Organize Their Own Memory?

Zichuan Fu, Junjia Qi, Wenlin Zhang, Yimin Deng, Shirong Wang, Jinyang Li, Xian Wu, Xiangyu Zhao

原文摘要

In most LLM agents, weights are fixed after deployment, so new information must be retained in external memory. Existing approaches often rely on predefined records and fixed organizational rules, which cannot readily accommodate evolving tasks and interactions. This raises the question: can LLM agents organize their own memory using ordinary documents? We introduce self-organized memory (SoMem), a framework that stores memory in linked documents. The agent decides what to summarize, how to group related information, and when to revise existing topics, and every summary it produces stays linked to the original interactions as evidence. The runtime validates updates and automatically derives temporal, relational, and recent-memory views from the organized content. At query time, the agent chooses what to search and read, following document links and inspecting original evidence as needed. On LoCoMo, BEAM-100K, and LongMemEval-S, SoMem outperforms the strongest baseline by 2.66, 3.10, and 2.00 percentage points, respectively. The constructed memory can be reused across models, and additional reasoning effort can further improve answer quality. These findings demonstrate the feasibility of agents organizing and using their own persistent memory through ordinary documents. Code is available at: https://github.com/Qi202/SoMem.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.11286

Cable construction effects on the lateral forces and wakes of helically stranded cables under oblique inflow

Tielin Ma, Junlin Zhang, Jingcheng Fu, Chongwen Jiang, Chuan Ye

原文摘要

Helically stranded cables in aerial towing systems experience oblique inflow, but are often modeled as smooth cylinders, neglecting three-dimensional cross-stream coupling and surface texture. We perform large-eddy simulations and spectral proper orthogonal decomposition for a smooth cylinder and four stranded cables at four inclinations from $22.5^\circ$ to $90^\circ$, all stationary, with $Re \approx 1.0 \times 10^4$. Aerodynamic differences are governed mainly by cross-sectional symmetry and strand-scale groove depth. Force-source decomposition shows that surface pressure contributes 94.9%--98.3% of the mean normal force and dominates the mean lateral force. High-symmetry, quasi-circular envelopes maintain balanced circumferential pressure, whereas deep-grooved, low-symmetry tri-lobed geometries produce asymmetric separation and a persistent lateral pressure imbalance. In the wake, high-symmetry cables sustain spanwise-synchronized Karman shedding at moderate-to-high inclinations, concentrating fluctuation energy in a dominant low-rank mode. Deep grooves disrupt shedding coherence, deflect the recirculation zone laterally, and spread energy across broadband scales. Mean lateral force and lateral-force fluctuations are therefore decoupled: the quasi-circular $(1 \times 19)$ cable has minimal mean lateral force but strong periodic fluctuations, while the deep-grooved $(3 \times 7)$ cable has weak fluctuations but appreciable mean lateral force. These results support construction-dependent cable selection and aerodynamic load modeling.

physics.flu-dyn · cs.CE · cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2608.22788

TailSieve: Partial-Rollout-Guided Tail Routing for LLM Rollouts

Tianqi Xu, Lu Lv, Haoyang Huang, Wenjie Huang, Zhanming Shen, Yuhao Shen, Baolin Zhang, Xinyi Hu 等 18 位作者

原文摘要

Large-scale rollouts have become a core component of modern LLM systems, spanning reinforcement learning (RL) post-training, on-policy distillation, and sampling-heavy evaluation. Unlike online serving, which typically targets request-level latency and throughput, synchronous rollouts must collect a complete batch before proceeding. A few long-tail generations can therefore dominate an entire rollout step, and uniform routing can slow these generations further by placing them in high-concurrency decoding batches. We present TailSieve, a partial-rollout-guided framework that jointly controls tail routing and replica allocation. Motivated by the observed persistence of prompt-level tail tendencies across policy updates, TailSieve uses partial-rollout completion status as a training-free signal for identifying candidate tail groups. Selected prompts are combined with newly sampled prompts, and all responses are generated from scratch under the current policy. A hierarchical controller uses response-work history and a serving model: a fast inner loop adjusts the number of isolated groups to balance completion times, while a slower outer loop adapts how many replicas serve these groups. The resulting low-concurrency execution also enables route-specialized speculative decoding with MTP or DFlash. Across the evaluated configurations, TailSieve achieves up to $1.67\times$ speedup with routing alone and $2.59\times$ with routing and speculation combined, both over uniform group routing.

cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.21613

To Think or Not to Think: Pre-Decisional Reasoning Budgets for Referring Audio-Visual Segmentation

Haojie Hu, Senda Chen, Ying Shen, Lin Zhang

原文摘要

Multimodal reasoning systems increasingly assume that longer chain-of-thought (CoT) uniformly improves downstream grounding. We show that this assumption fails in referring audio-visual segmentation (Ref-AVS): simple queries can be harmed by excessive reasoning, creating an "overthinking trap" associated with phrase and localization drift, while genuinely ambiguous queries still benefit from multi-step reasoning. To systematically study this, we construct counterfactual reasoning-budget labels under a five-fold out-of-fold evaluation protocol. Our results recast adaptive Ref-AVS reasoning as a pre-decisional representation problem: before the first generated token, the generation-onset state contains linearly decodable information predictive of whether additional reasoning will improve segmentation. A linear probe predicts binary reasoning need with 70.1% accuracy, substantially outperforming text-only features and uncertainty-based proxy signals. As a router, the onset-state controller improves over downstream confidence routing and closely matches Short-then-Decide while using far fewer tokens. Proxy comparisons show that the onset representation contains predictive information not fully recovered by the evaluated hand-crafted features. Reusing this onset state for lightweight, single forward-pass routing reduces autoregressive token consumption by 60% while retaining about 96% of always-long performance. It further provides a cost-quality trade-off on $R^2$-AVSBench without target-label tuning, rather than an accuracy improvement over always-long. Error analysis links harmful long reasoning to phrase and localization drift, illustrating the benefit of bypassing unnecessary reasoning on simple queries. Ultimately, our results support efficient adaptive reasoning in Ref-AVS through pre-decisional internal representations.

cs.MM

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.06522

Anatomy of LLM Sycophancy: What a Flip Rate Hides

Haonan Huang, Tianlin Zhang

原文摘要

A model under pushback can correct itself, capitulate, or hold, and one flip rate counts a correction and a capitulation alike. SycoLens, a modular replay protocol, measures flip rates from replays of an item, a committed answer, and one scripted user line, each against a matched control with the line deleted. Pushback wording, committed text, answer format, boundary distance, and ground truth become factors of one instrument, where earlier instruments vary one to three, so the interactions between the factors become measurable. Across eleven frontier models and about 760,000 controlled replays, lines that assert the opposite verdict and lines that challenge without asserting one rank the models almost unrelatedly. Flip effects grow several-fold near a model's boundary, yet items the model is certain of still carry about half of the most-affected totals; planted prose anchors a commitment at rest more than it protects it. Under a yes/no readout the rankings come closer, entangled with a pressure-induced shift toward "no". Where a truth exists, one model re-derives and corrects itself while another abandons correct answers without written work. One score per model therefore compares different behaviours; we condense them into a reporting profile, and the records and analyses will be released upon publication.

cs.CL · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.08566

CascadeRec: Deployed Adaptive LLM Compute Allocation for Multi-Tenant Content Recommendation

Kai Zhao, Guilin Zhang, Xiang Cheng, Xiquan Cui, Henry Heng, Xu Chu, Alet Blanken

原文摘要

Large language models (LLMs) are attractive rerankers for recommendation. Running an expensive model on every request is costly, and for users with sparse interaction histories it often lowers quality. We present CascadeRec, a two-tier LLM reranking cascade deployed in production at a large financial planning and analysis (FP&A) SaaS provider. It serves content recommendations to enterprise tenants on AWS EKS and GKE. A lightweight Tier-1 model reranks every request. A router escalates only the requests where extra inference-time compute is likely to help, and sends them to a more expensive Tier-2 procedure (self-consistency sampling, multi-path reasoning, or a larger model). The system has run in a production A/B experiment for over a month. Compared with uniform use of the expensive tier, CascadeRec has reduced inference cost by about 10%, with minimal latency impact and no quality regression observed to date. Tenant data cannot be released, so we also evaluate the pipeline on public benchmarks (MovieLens-1M and Amazon-Electronics). Segment-level analysis with paired-bootstrap confidence intervals shows that uniform LLM reranking lowers NDCG@10 for sparse-history users by up to 30%, and its only statistically significant gain appears on the segment with the longest histories. Escalating cold users can make the loss larger instead of repairing it. Which segments benefit from escalation depends on the domain, and this motivates the segment-gated rollout we use in production. We describe the deployment architecture, the measured benefits, and lessons learned, including inter-tenant load imbalance as a driver for cascading and the handling of LLM output-parsing failures at the serving layer.

cs.IR · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.06207

Taming the Entropy Cliff: Variable Codebook Size Quantization for Autoregressive Visual Generation

Bowen Zheng, Weijian Luo, Guang Yang, Colin Zhang, Tianyang Hu

原文摘要

Most discrete visual tokenizers rely on a default design: every position in the sequence shares the same codebook. Such a constant-codebook design can consume the finite-data entropy budget very quickly. We observe that the per-position empirical conditional entropy decays so quickly along the sequence that, after a few positions, the conditional distribution becomes essentially deterministic. On ImageNet with $K=16384$, this happens after only about 2 out of 256 positions, when most prefixes already identify individual images in the dataset. We call this phenomenon the Entropy Cliff. Under full conditional utilization, larger codebooks exhaust the finite-data entropy budget in fewer positions. The empirical entropy decreases more gradually in our language comparison, where the effective entropy per position stays below the codebook capacity. To address this, we propose Variable Codebook Size Quantization (VCQ), where the codebook size $K_t$ grows monotonically along the sequence from $K_{\min}=2$ to $K_{\max}$, leaving the loss function, parameter count, and AR training procedure unchanged. With a vanilla autoregressive Transformer and standard next-token prediction, VCQ-Base achieves gFID w/o CFG 14.80 on ImageNet $256\times256$, compared with 27.98 and 18.16 for constant codebooks of size 16384 and 8192, respectively. Scaled up, it reaches gFID 1.71 with 684M autoregressive parameters, without any extra training techniques such as semantic regularization or causal alignment. The extreme information bottleneck at $K_{\min}=2$ encourages semantic concentration in early encoder features: a linear probe on features from only the first 10 positions reaches 43.8% top-1 accuracy on ImageNet, compared to 27.1% for uniform codebooks. These results show that what matters is not only the total capacity of the codebook, but also how that capacity is distributed and organized.

cs.CV · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06982

Decode Before You Hear: Speculative Listening for Chunk-Interleaved Full-Duplex Speech Language Models

Yutong Bian, Yanglin Zhang, Yuxiang Wang, Luoyuan Zhang, Rongshen He, Zihao Wang, Junwen Qiu, Jiaqi Li 等 9 位作者

原文摘要

Chunk-interleaved full-duplex speech language models listen and speak in fixed time chunks, and each chunk must be processed before the next one arrives. Their workload is asymmetric across chunks: listen chunks leave compute idle while awaiting audio, and speak and tool-call chunks require serial generation, which delays responses and tool calls. Using the idle compute for the next chunk requires decoding it before its audio arrives, under an assumed input. Taking silence as the assumed input and comparing the output with the output under real audio on four models, we observe two properties. First, the next audio chunk does not change the output in 95.3% of chunks. Second, 88% of changes begin at the first output token. Building on these properties, we propose Speculative Listening, a training-free, lossless framework for input-time self-speculative decoding. During idle time, the model drafts the next chunk under silence and, for changes at the first token, a second branch from the runner-up first token. It keeps all drafts in a cross-chunk hypothesis pool and verifies them when the audio arrives. On our eight-task benchmark TickTock-Bench, Speculative Listening speeds up speak chunks by 2.6-2.8x and tool-call chunks by up to 23.4x. On average, it reduces the decoding delay before a response starts by 73%, to 74-241 ms, close to the 200 ms gap between human turns, and the maximum lag behind real time by 94.4%.

cs.SD

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2607.05721

SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation

Yimeng Zhang, Yingying Zhuang, Ziyi Wang, Yuxuan Lu, Pei Chen, Aman Gupta, Zhe Su, Ming Tan 等 15 位作者

原文摘要

Uncertainty estimation is essential for the trustworthy deployment of LLMs: it underpins self-refinement and serves as a decision signal for downstream systems, from verifiers to collaborating agents. However, existing approaches operate at suboptimal granularities: token-level scores lack semantic coherence, while sequence-level scores fail to localize errors. We formalize Span-Level Uncertainty Estimation (SLUE), a new task that targets the natural granularity for uncertainty: semantically coherent text spans, each conveying a single assessable unit of meaning. To address this task, we introduce SPANUQ, a lightweight probe that distills the uncertainty knowledge from expensive multi-sample inference into a single forward pass over LLM hidden states. SPANUQ employs a DETR-style span decoder to simultaneously detect spans and estimate their uncertainty via a Mixture of Beta distribution, trained with a principled combination of Beta NLL regression and contrastive ranking objectives. We construct SPANUQ-BENCH, the first span-level uncertainty benchmark comprising 20K prompts, 293K annotated spans, and continuous soft labels derived from multi-sample claim verification. Experiments on five LLM backbones show that SPANUQ consistently achieves the best span-level uncertainty quality , outperforming the strongest probe baseline and all sampling-based methods while being 10~20x faster. Its DETR-based span detector attains 0.910 F1, surpassing the best heuristic by 39.4%, enabling precise error localization that sequence-level methods cannot provide. The same architecture ports to five LLMs spanning two model families, with one probe trained per backbone, and we additionally observe that sequence-level uncertainty is partially decomposable, suggesting that span-level estimation subsumes sequence-level as a special case. The project page is available damon-demon.github.io/SpanUQ.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04897

TapeSAC: Automated, Pareto-Optimal Fine-Grained Memory Management for Large Model Training

Kaihui Ren, Yifan Zhao, Zejian Wang, Yingtong Xiong, Xiaoling Zhou, Daohong Xu, Chenhao Guan, Zhikang Mo 等 14 位作者

原文摘要

Selective activation checkpointing (SAC) decides which activations to save for the backward pass and which to recompute: each choice is a point in the plane of peak memory and recomputation time, and the goal is that plane's Pareto frontier. Current practice falls short of this goal. Hand-tuned policies fix a single point that does not adapt when the model, batch or kernels change; PyTorch's SAC reruns each block's forward pass in program order, recomputing values the backward pass never reads; and the memory budget of torch.compile treats recomputation as free or forbidden and selects saved tensors by bytes alone, an assumption that fails for the custom kernels that produce many of a large model's activations. We present TapeSAC, which automates SAC in eager PyTorch without editing the model's forward code. TapeSAC records each block's forward pass at the dispatcher and, during backward, replays only the producers of the values autograd consumes, so saving a tensor eliminates every producer needed only for it. A planner selects the saved tensors from measured recomputation costs under a single limit on step-peak memory. On Qwen3.8-27B and Qwen3.6-35B-A3B, replay alone reduces step time by 3.0% and 1.7% versus full checkpointing at no greater memory cost; memory-aware planning raises these gains to 6.3% and 13.9% under a 90%-of-device-memory limit. The evaluated hand-tuned policies fall short of this memory-recomputation frontier and, on Qwen3.6-35B-A3B, run out of memory at larger batches. Over 500 steps on real data, TapeSAC reduces training time by 8-11%; on Qwen3.5-397B, it is 2.7% faster than a model-specific hand-tuned policy while using 14% less memory.

cs.DC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04834

Feeling of Knowing in Large Language Models

Zichuan Fu, Xian Wu, Jingtong Gao, Wenlin Zhang, Jiaxuan Li, Binhao Wang, Yimin Deng, Guojing Li 等 12 位作者

原文摘要

Post-training updates change what large language models (LLMs) can answer, but may leave their answerability judgments misaligned. We study this knowledge-judgment desynchronization through functional feeling of knowing (FoK), which supports selective generation, retrieval triggering, and model routing without implying subjective experience. We propose two trainable channels: a knowledge channel that changes answerability and a FoK channel that judges the current knowledge state. Three stages train them: on-policy supervised fine-tuning (OP-SFT) updates knowledge from corrected on-policy answers; knowledge-delta FoK training (K$Δ$-FT) adapts judgments across intermediate knowledge states; and FoK-guided policy optimization (FGPO) trains the subsequent answer or explanation. Factual-QA experiments, cross-dataset evaluation, and three sequential updates show that explicit FoK alignment improves judgment quality on updated knowledge. MMLU and ARC-Easy separately assess capability preservation.

cs.AI · cs.CL · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03228

AnchorComplete: Recovering Complete Scene Meshes from Partial Point Clouds

Weirong Chen, Ismail El-Yamany, Ganlin Zhang, Chuanxia Zheng, Andrea Vedaldi, Daniel Cremers

原文摘要

Most image-based 3D reconstructors primarily recover visible surfaces, producing partial point clouds with large holes and leaving unobserved geometry unresolved. To address this limitation, we introduce AnchorComplete, a framework that reconstructs complete scene meshes from partial point clouds while preserving observed geometry and supporting scenes of varying scales. Building on TRELLIS.2, our method combines partial-point conditioning, which integrates global scene tokens with local spatial evidence to anchor generation to the input, and metric-scale coarse-to-fine generation, which progressively refines scene geometry within fixed-size windows while maintaining global structural consistency. Experiments on Replica and SCRREAM demonstrate that AnchorComplete consistently outperforms strong baselines in complete-scene reconstruction, particularly under sparse-view settings.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.21312

Spectrum Is Not Enough: A Random-Matrix Account of Data-Free Preconditioning for Differentially Private Learning

Shilin Zhang, Kangning Wang

原文摘要

At a fixed privacy budget, curvature is exactly the object the budget makes unavailable. A recent answer estimates it from synthetic probes, at neither privacy cost nor distribution shift, and validates a probe by the eigenvalue decay it reproduces. That test reads the spectrum, not the basis, and what a probe reconstructs is unknown. This paper gives a random-matrix account of data-free preconditioning, organized by one dichotomy. Of the two Kronecker factors, the error-signal factor $ G $ and the architecture-determined bulk of the activation factor $ A $ are reconstructible without data up to an explicit error-energy bound; only the data-induced spikes of $ A $ are not. We verify that bound with zero violations along a full training trajectory, though it stays one to nearly three orders of magnitude above the gap it bounds. We locate the failure boundary at a Baik-Ben Arous-Péché threshold: a bulk-only probe suffers an exact damage $s^2/(1-s^2)$ from a spike of strength $s^2$, invisible once its amplitude exceeds the scaled-bulk threshold $( \sqrt {c}/(1+ \sqrt {c}))^{1/2}$, $c = d/n$. The test in use is basis-blind: rotating only the probe eigenbasis at a near-exact matched spectrum inflates the residual by $2.3 \times $ to $89.8 \times $, and a closed-form volume law fixes the within-spectrum ordering of the critical volume while a two-oracle construction proves that no functional of the probe spectrum fixes its magnitude. Over forty seeds on three modalities, per-layer tuning does not transfer to accuracy, no data-free arm beats the flat probe, public-data curvature does on both modalities tested, and the ordering inverts on PathMNIST under real DP noise. The account yields a probe-design protocol and falsifiable predictions for the regimes where data-free curvature can be trusted.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02846

SpecOPD: Branch-Parallel On-Policy Distillation for Speculative Decoding

Tianyu Liu, Jun Zhang, Wenjie Huang, Yuhao Shen, Tianqi Xu, Xinyue Hao, Hongyaoxing Gu, Xuyang Liu 等 15 位作者

原文摘要

Speculative decoding (SD) accelerates large language model inference by reducing the sequential cost of token-by-token generation. Advanced SD methods reduce proposal cost with parallel drafters that predict entire blocks in one forward pass, but overall speedup still depends on acceptance length. On-policy distillation (OPD) can push well-trained drafters beyond acceptance-length plateaus, yet conventional pipelines collect supervision through sequential draft-and-verify rollouts. Our key insight is that on-policy supervision need not inherit these dependencies: a draft block can be supervised using only its reference prefix and draft-sampled history. We propose SpecOPD, a branch-parallel OPD framework that parallelizes draft sampling and target supervision across independent branches drawn from the evolving drafter policy at reference anchors. We pack these branches into proposal trees to share target-prefix computation, using tree masks for full attention, branch-consistent indexer groups for sparse attention, and indexed state replay for linear attention to preserve each branch's history. We train the drafter with an E2E-TV surrogate that directly targets acceptance length. Our study spans 8B full-attention (GQA), 27B full-linear (GQA--GDN), and 125B sparse-linear (QSA--GDN) targets, three parallel draft architectures (DFlash, DSpark, and DFlash V2), and four task domains (dialogue, coding, mathematics, and agentic). SpecOPD achieves up to 41.45% higher mean acceptance length, $16.09\times$ supervised-token throughput over sequential-rollout OPD under native configurations, and up to $3.56\times$ generation speedup over autoregressive decoding.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2607.17281

AIGB-R1: Self-Evolving Generative Auto-Bidding via Hierarchical Planner-Executor Optimization

Yuejia Dou, Hesong Wang, Xinyu Zhang, Yuchao Ma, Tianyu Wang, Zhilin Zhang, Chuan Yu, Jian Xu 等 10 位作者

原文摘要

Auto-bidding plays an essential role in online advertising, automatically adjusting bids for advertisers to optimize their commercial goals. The emerging AI-Generated Bidding (AIGB) paradigm widely adopts generative modeling to optimize bidding strategies, yet suffers from the limited mode coverage of offline datasets and inadequate task-state understanding, hindering effective exploration of optimal strategies. Large Language Models (LLMs), with prior world knowledge and reasoning capabilities, offer a promising approach to overcome these limitations. However, directly applying LLMs to auto-bidding tasks faces inherent challenges in limited numerical precision, hallucinations, and inference latency. To address these limitations, we propose AIGB-R1, a hierarchical self-evolving auto-bidding framework aiming to enhance AI-Generated Bidding via LLMs' Reasoning capabilities, comprising a high-level Planner module for macro-level strategy planning and a low-level Executor module for fine-grained decision-making. Building upon this, we design an experience-driven self-evolving loop, enabling autonomous strategy exploration and optimization from accumulated experience. We adopt a two-stage pipeline of offline pre-training and post-training alignment, and build an interactive bidding simulation environment for strategy rollout. Furthermore, we propose Decoupled Group Relative Policy Optimization (D-GRPO) to achieve end-to-end optimization via advantage decoupling. Experimental results on a large-scale public dataset demonstrate the effectiveness of AIGB-R1.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02593

AVID: A 5T fMRI Dataset for Benchmarking Auditory-induced Visual Mental Imagery Decoding

Shiqi Shen, Shurui Li, Xilin Zhang, Yuanning Li

原文摘要

Visual mental imagery (VMI) decoding uses brain activity to reconstruct internally generated visual scenes, providing a potential pathway for externalizing internal experiences and developing neuroprosthetic communication interfaces. However, progress is limited by the lack of dedicated, training-scale VMI datasets. Current VMI paradigms often rely on visual prompts or short-term memory images, introducing visual and mnemonic confounds that make it difficult to disentangle neural activity associated with endogenous image construction from signals driven by prior or concurrent visual input. Furthermore, existing VMI datasets offer limited descriptive control over individual visual attributes, making it difficult to systematically assess fine grained visual details. To address these gaps, we introduce AVID, the first dedicated training-scale 5.0 Tesla functional magnetic resonance imaging (fMRI) resource for auditory-induced VMI decoding. AVID comprises approximately 67 hours of dense recordings from nine participants. The dataset is structured into two complementary splits: a scene-level complex split using naturalistic Mandarin descriptions and a controlled simple split with explicitly specified visual attributes. We further formalize a standardized evaluation protocol that decouples source-image consistency from semantic alignment, allowing for a more nuanced assessment of imagery reconstruction. Baseline qualitative and quantitative results show that current reconstruction pipelines remain limited on AVID, although the most successful reconstructions preserve recognizable scene level semantics from spoken descriptions. By combining dedicated training-scale neural recordings and standardized evaluation resources, AVID provides both a data foundation for imagery-specific model development and a benchmark for evaluating VMI decoding with substantially reduced sensory contamination.

cs.CV · cs.AI · q-bio.NC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02463

Think Globally, Generate Personally: Multi-View Thinking with Adaptive Generation for Generative Recommendation

Yuxuan Hu, Yingyi Zhang, Wenlin Zhang, Chao Zhang, Haotian Nan, Ziwei Liu, Jiancheng Dong, Tongzhou Wu 等 14 位作者

原文摘要

Semantic-ID-based (SID) generative recommendation (GR) represents each item as a sequence of discrete tokens and predicts the next item through autoregressive SID generation. However, existing methods face two challenges. First, insufficient shared context modeling for user sequences: full-history and recent behaviors provide overlapping yet complementary evidence, but modeling them within a single sequence may dilute current cues, whereas processing them through separate streams may fail to coordinate them. Second, insufficient personalized adaptation in progressive SID decisions: although decoder states evolve with generated prefixes, existing models typically use a shared output-side transformation across recommendation instances, without explicitly selecting a transformation suited to the current user need. To address these challenges, we propose Multi-View Thinking with Adaptive Generation (MV-Thinker), a framework that Thinks Globally and Generates Personally. First, Multi-View Temporal Thinking uses view-specific thinking queries and readers over overlapping full-history and recent-history views, coordinating both sources into complementary user-conditioned contexts before SID generation. Second, Hierarchical Residual Adaptation organizes lightweight adapters into nested residual paths before a shared vocabulary projection, adaptively transforming context-bearing decoder states for SID decisions. Extensive experiments on four public datasets demonstrate that MV-Thinker consistently outperforms strong discriminative and generative recommendation baselines across multiple ranking metrics. The implementation code and datasets are available on: https://anonymous.4open.science/r/MV-Thinker

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02365

SCARF: A Scene-Adaptive Depth-Guided G-3DGS Encoder Accelerator with Semantic Reuse and Fused Dataflow

Zirui Ma, Zhihua Fan, Haibin Wu, Wenxing Li, Fulin Zhang, Dongrui Fan, Xiaochun Ye, Wenming Li

原文摘要

Depth-guided, cost-volume-based Generalizable 3D Gaussian Splatting (G-3DGS) encoders reconstruct scenes as sets of Gaussian primitives directly from multi-view images without requiring per-scene optimization, thereby facilitating feed-forward novel-view synthesis on edge computing platforms. However, their shared encoder architecture presents two primary bottlenecks. First, the per-pixel depth search disregards semantic similarity within the 2D feature space, causing depth search operations to account for over 65% of memory traffic due to redundant exhaustive candidate scans. Second, the sequential pipeline of depth prediction followed by Gaussian generation neglects local 3D geometric continuity, resulting in more than 52% of generated Gaussians being redundant prior to instantiation. In this paper, we introduce SCARF, a dataflow-co-designed accelerator that aligns each bottleneck with a corresponding scene-level regularity. To mitigate the first bottleneck, Feature Similarity Depth Reuse reconceptualizes depth search as a problem of semantic locality: depth consistency is determined by affinity in feature space rather than physical address proximity. It transforms exhaustive per-pixel depth searches into anchor-centered local searches. To overcome the second bottleneck, Scene-Adaptive Early Sparsification disrupts the conventional serial depth-to-materialization boundary by probing each tile to infer local scene regularity. It then selectively activates only the minimally sufficient generation pathway, eliminating redundant Gaussian computations before primitive instantiation. On a 27.8 square millimeter 28 nm post-layout design, SCARF delivers a 2.94x geometric-mean end-to-end speedup with maximum PSNR degradation below 0.072%, 2.23 inferences per joule, and 34.5x GPU energy efficiency.

cs.AR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.01937

Spade-SLAM: Sparse Correspondence Anchors for Dense Self-Supervised Visual SLAM

Julian Philipp Schmitt, Weirong Chen, Ganlin Zhang, Felix Wimbauer, Daniel Cremers

原文摘要

We introduce Spade-SLAM, a self-supervised dense visual SLAM framework that substantially narrows the long-standing performance gap to supervised methods in a zero-shot setting. By demonstrating that high-fidelity tracking can be achieved without explicit ground-truth pose and depth labels, our approach offers a highly effective and label-efficient alternative to fully supervised pretraining. Our core contribution is a complementary sparse-to-dense training regime: While a warping-based photometric loss provides a dense training signal, we stabilize the optimization and suppress drift in ambiguous regions using a sparse correspondence loss derived from image feature matches. Extensive experiments demonstrate that Spade-SLAM significantly improves robustness and accuracy compared to existing self-supervised baselines, closely rivaling SLAM systems trained with explicit pose and depth supervision. The code will be made publicly available at https://github.com/JulianPSchmitt/Spade-SLAM.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.01840

Self-Evolving Agentic Workflows via Semantic Decomposition of Complex Tasks

Zichuan Fu, Junjia Qi, Xian Wu, Wenlin Zhang, Yimin Deng, Kaifeng Guo, Xiangyu Zhao

原文摘要

Agentic workflows organize complex tasks into reusable procedures. However, invoking an LLM at every step incurs repeated inference costs and delays. We propose a framework for self-evolving agentic workflows through semantic task decomposition. Each subtask has explicit applicability conditions and a procedure that executes code or invokes a model as needed. An agent uses execution feedback to refine the decomposition and update these procedures. Experiments in the GLEE competition show the framework on bargaining, negotiation and persuasion. Most observed final-decision procedures are implemented in code. Code is available at https://github.com/Applied-Machine-Learning-Lab/NeurIPS2026_SeaFlow.

cs.AI · cs.CL · cs.MA

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.40166

EventSeeColor: Learning to Reconstruct Color from Monochrome Events

Chuanzhi Xu, Langyi Chen, Haoxian Zhou, Zelin Zhang, Haodong Chen, Qiang Qu, Xiaoming Chen, Weidong Cai

原文摘要

Event cameras capture visual brightness changes under challenging lighting conditions. However, traditional event-based image reconstruction targets grayscale intensity and produces images without scene color. Color distinguishes surfaces with similar intensity. Reconstructing color from events can reveal appearance differences that reconstructed grayscale images cannot express. Although monochrome event cameras do not directly measure color, their event-triggering patterns depend on surface reflectance, illumination, and the sensor's spectral response. These patterns may contain cues that support color reconstruction. In this paper, we present EventSeeColor, which uses event-reconstructed grayscale and aligned events to jointly guide a diffusion colorizer. Grayscale-derived semantic context supports color inference, while region-balanced chroma supervision and local RGB refinement improve color distribution, brightness, and texture. In the combined evaluation on two real event datasets, EventSeeColor achieves the best structural similarity, learned perceptual scores, and chromatic-region color difference among the external methods.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗