Humanoid rigging for modern characters with complex garments remains challenging due to the need for dense auxiliary skeletons and precise skinning control. Existing automatic methods primarily recover rigs from geometry, providing limited artist control over auxiliary articulation and often lacking explicit semantic or kinematic-chain structure for dense garment bones, leading to instability and weight bleeding. We present TailorRig, a framework that reformulates garment-aware humanoid rigging as a structured, controllable full-rig generation problem. This formulation shifts the goal from automatic rig recovery to artist-driven rig authoring, where both body-template compatibility and auxiliary articulation are explicit design choices. TailorRig exposes two explicit axes of artist control within a unified framework: a user-specified body skeleton template that preserves body-rig compatibility, and artist-provided semantic labels that drive an autoregressive generator to synthesize requested category-specific auxiliary bone chains. To robustly assign skinning weights under dense skeletal configurations, we further introduce a hierarchical chain-to-bone formulation: vertices are first associated to a kinematic chain, and bone weights are then refined within that chain, reducing weight bleeding between spatially adjacent but semantically unrelated structures. We evaluate TailorRig on a diverse set of characters and introduce Human-Mix, a large-scale dataset of over 38,000 humanoid models curated for garment-aware rigging. Experiments show that TailorRig improves skeleton accuracy, deformation stability, controllability, and practical usability over existing methods. Project page: https://xxxx0601.github.io/trig/ ; code: https://github.com/Tencent/TailorRig ; dataset: https://huggingface.co/datasets/tencent/TailorRig
Enhan Zhao, Yongqi Tong, Xuejun Yu, Hang Wang, Pan Wang, Yichao Wang, Di He, Jianshe Li 等 10 位作者
Existing long-horizon agent benchmarks often emphasize final task success while providing limited assessment of intermediate execution validity. Yet real-world business operations require agents to navigate extensive, potentially outdated or conflicting policies while satisfying critical constraints throughout execution. These requirements are particularly consequential in high-stakes domains such as payments, where intermediate violations can incur financial losses or regulatory consequences regardless of the final outcome. To bridge this evaluation gap, we introduce BizAgentBench, a multi-turn benchmark grounded in real-world procedures from a leading payments company. Its executable environment integrates 789 knowledge documents and 601 discoverable business tools, with consistent policies, states, and tool behavior. The benchmark comprises 50 tasks involving resource constraints, temporal dependencies, prior commitments, and conflicting or outdated guidance. Each task is validated through reference execution, while deterministic checks assess both goal attainment and execution validity without prescribing a single solution path. Experiments reveal substantial limitations even among frontier models: Claude Opus 5 and GPT-6 Sol achieve strict success rates of only 64.0% and 61.3%, respectively. For Qwen3.8-27B, goal attainment reaches 42.0%, but strict success drops to 32.7%, demonstrating how outcome-based evaluation can mask invalid execution. Beyond success rates, DeepSeek-V4.1-Flash averages 191.3 tool calls per run, approximately 70% more than reference executions, revealing substantial execution overhead. Overall, BizAgentBench establishes a testbed for evaluating and advancing long-horizon agents beyond task completion, toward reliable and constraint-compliant execution in complex business environments.
Chen Huang, Jingbo Wang, Zhemin Zhang, Ming Zhong, Zhuo Fu, Zhiding Liang, Yuan Sun, Dong E. Liu
Neutral-atom quantum computers commonly connect distant qubits by transporting atoms, incurring trap-handover errors, heating, and atom loss. We present BRIDGE (Buffer-Relay Interconnect for Data-stable Gate Execution), a hardware--software co-design that connects stationary data atoms through a fixed, second-species buffer network. Short buffer chains mediate remote controlled-Z gates and return to their initial states after each operation. BRIDGE-F combines relay routing with buffer-mediated SWAPs to relocate quantum states without transporting atoms. BRIDGE-H additionally permits data-atom transport when its full modeled cost is offset by higher survival-weighted fidelity. Across 21 circuits, BRIDGE-F exceeds ZAP and Enola in modeled survival-weighted fidelity, with geometric-mean ratios of $2.1\times$ and $38\times$ and medians of $1.3\times$ and $3.7\times$. Modeled execution latency is lower by factors of approximately $500$ and $1{,}350$, respectively, with circuit single-qubit costs excluded and relay/SWAP Hadamards included. On the same fabric, BRIDGE-F matches or exceeds SABRE-based routing on 17 circuits. Rescoring nominal plans with relay-module error equal to direct-CZ error preserves the advantage over ZAP across all 21 circuits, although recompilation can underperform on QFT. These results identify stationary buffer-mediated connectivity as a promising alternative to repeated atom transport under the stated physical assumptions.
Fazong Wu, Xin Wang, Ming Yang, Zhenyong Zhang, Xiaoming Wu
As large language models (LLMs) are increasingly deployed as general-purpose conversational assistants, they must handle diverse user requests while reliably refusing harmful ones. Jailbreak attacks craft adversarial instructions or conversational contexts to elicit unsafe responses despite these safeguards. However, existing jailbreak benchmarks often rely on manually designed or collected prompts, making it difficult to systematically extend coverage to new threat conditions. Moreover, existing attack search methods typically optimize for attack success rate (ASR), providing limited insight into how model behavior changes locally between safe refusal and unsafe response. To address these limitations, we present JailBound, a unified jailbreak evaluation framework that combines automated benchmark construction with first-order loss (FOL)-guided dual-branch search. JailBound organizes threats by risk category, application domain, and attack type. Conditioned on this taxonomy, a fine-tuned meta-attack generator synthesizes evaluation prompts across diverse threat conditions. FOL serves as a local sensitivity signal for exploring high-value attack states and behavior-sensitive states. We evaluate JailBound across 46 LLMs from 13 model families under a unified experimental protocol. Under matched settings, the High-Value Branch improves average ASR over a ZOL-only baseline by 5.1 and 4.9 percentage points across the two evaluation settings. Under local perturbations, states from the Safety-Sensitivity Branch exhibit higher behavioral flip rates than those from the High-Value Branch across all tested radii for JailBound and S-Eval, indicating greater local sensitivity of judge-assigned safety labels. Warning: this paper includes examples that may be offensive or harmful.
Miner extractable value (MEV) in automated market makers allows block builders to profit from transaction ordering and injected trades, imposing costs on users and contributing to builder centralization. We introduce Otter (Optimal Truthful Trading with Excess Redistribution), a two-asset batch AMM that achieves provable MEV resilience when the consensus layer provides censorship resilience and block space is uncongested. Otter makes truthful behavior a dominant strategy for both users and builders. Consequently, a builder cannot profit from strategic deviations, including reordering bids or injecting sybil bids, even when the builder is itself a trader with intrinsic value. Moreover, we show that Otter maximizes social welfare, in a strong sense, within a natural class of mechanisms satisfying the desired game-theoretic properties. To achieve these guarantees, we introduce a new paradigm called surplus redistribution, which provably prevents residual surplus from being captured as MEV by redirecting it to the broader community. Specifically, the pool's output tokens need not be distributed entirely among users in the current batch. Instead, residual surplus may be transferred to a smart contract governed by the decentralized community. The accumulated surplus can be used to benefit community members --- for example, by subsidizing transaction fees, rewarding liquidity providers, or returning assets to the pool. Our approach relies on the underlying consensus layer to provide censorship resilience. We motivate its necessity through an impossibility result showing that the desired game-theoretic guarantees become unattainable when the builder is additionally allowed to censor transactions. Thus, our results also provide a mathematically formal demonstration of how consensus-level security guarantees can fundamentally expand what is achievable at the application layer.
Large language models (LLMs) are increasingly recognized for their remarkable capabilities; federated learning (FL) offers a promising framework for collaboratively fine-tuning LLMs on edge devices while achieving data privacy. However, edge devices are inherently heterogeneous: clients may exhibit diverse data distributions and resource constraints, including limited memory and computation budgets. Recent federated Mixture-of-Experts (MoE) approaches address these challenges by using sparse expert activation or partial expert loading. However, existing partial-expert methods rely on heuristic full-pool observations or prioritize system-wide load balance, without guarantees for recovering client-specific relevant experts from partial observations. We propose $S^3$-MoE, a server-side sparse expert assignment framework that estimates client-specific expert relevance from partial observations and refines assignments through an exploit--explore mechanism. We also establish relevance-estimation error and exact Top-$K$ recovery guarantees. Experiments using a CNN-based MoE on CIFAR-100 show that $S^3$-MoE recovers task-dependent client--expert preferences and reduces conflicts caused by correlated expert selection across clients. Compared with Full-20 FedAvg, which gives each client access to all 20 experts for local Top-5 routing, $S^3$-MoE achieves 1.31 percentage points higher mean final accuracy while loading only $5$ experts per client.
Language-conditioned trajectory generation is here, but its evaluation has not kept pace. Existing pedestrian trajectory metrics compare trajectories with real-world human data. This does not scale to text-to-trajectory generation across diverse contexts, as collecting human trajectories for every scenario is costly and infeasible. Moreover, pedestrian behavior is heterogeneous and context-dependent, with no single metric as the correct answer, and current evaluation frameworks are not transferable to this domain. These challenges make scalable, reliable evaluation difficult. We introduce STRIDE, the first framework for evaluating context alignment between scenario descriptions and pedestrian trajectories. STRIDE addresses these challenges through three design choices. First, we derive our VRDST evaluation protocol from sociological theories to define a complete evaluation space. Second, it decomposes high-level context into scenario-adaptive behavioral questions. Third, every question is resolved against a deterministic measurement tool library that yields reproducible answers. Together, STRIDE enables verifiable, automated, and scalable evaluation across diverse contexts without requiring human trajectory data. We instantiate STRIDE in the crowd domain as STRIDE-Bench, comprising 1K scenarios, 6K behavioral questions, and 11K measurements with calibrated expected answers across 30 real-world maps. Comprehensive human validations show that STRIDE-Bench is consistent with human behavior and judgment, achieving 80% human agreement. We further evaluate several text-to-trajectory models, finding limited context-alignment capability and persistent challenges in fine-grained context conditioning. We believe that the STRIDE framework provides a first step toward principled evaluation of context-aligned pedestrian trajectory generation.
Accurate electricity load forecasting provides critical decision support for energy procurement, generation scheduling, and supply-demand balancing, contributing to the economic efficiency and operational reliability of power systems. However, future electricity demand can be affected by the complex interplay of multiple events described in Web notices, weather alerts, and event schedules. Their effects vary over time and can reinforce or offset one another, making electricity load difficult to forecast accurately. While existing methods can incorporate context as additional input, how to use the effects of multiple events and their interactions to guide the load forecasting process itself remains to be explored. To address this gap, we propose ElecLoad, a context-aware agent for multi-event electricity load forecasting that adaptively organizes the forecasting process. Specifically, we train ElecLoad to combine historical load with context to analyze each event's relevance to the target load, the timing of its effects, and potential interactions among events. Guided by this analysis, ElecLoad adaptively chooses to forecast the entire forecast horizon directly or to partition it into segments according to changes in event effects. When the horizon is partitioned, sub-agents independently generate initial forecasts for each segment. The main agent then reviews these forecasts, and the sub-agents perform selective refinement guided by its feedback. Experiments on multiple electricity load datasets show that ElecLoad achieves state-of-the-art performance in most evaluation scenarios. Code is available at https://github.com/zhjai/ElecLoad.
Spiking Neural Networks offer event-driven, energy-efficient computation for speech command recognition, yet most existing methods follow fixed computation paths, allocating redundant computation to easy samples while providing limited insight into how evidence evolves during inference. To address these challenges, we introduce SpikCTU, a compact spiking network with adaptive recurrent computation for speech command recognition. Specifically, a hybrid spiking embedding module preserves continuous input information alongside spike features to construct a fixed temporal memory. A shared Continuous Thinking Unit (CTU) repeatedly reads this memory using keys derived from spikes and continuous values, and updates a binary recurrent state with lightweight stage adapters. Furthermore, combining a fixed base prediction with accumulated corrections at each step enables learned aggregation over all steps or confidence-based early exit using only the predictions available so far. Experiments on three speech command recognition benchmarks show that SpikCTU achieves competitive accuracy with only 0.16--0.17M parameters, and the early exit strategy can substantially further reduce computational cost without compromising accuracy. In addition, trajectory and attention analyses reveal that recurrent steps enable error correction and progressively refine predictions. Code is available at https://github.com/Rivflyyy/spikctu.
As large language models (LLMs) are increasingly applied for emotional companionship, evaluating their behavior and capabilities in intimate relationships has become a pressing issue. However, existing assessments primarily characterize general personality traits, providing limited insight into model behavior within intimate and emotionally sensitive contexts. Therefore, we introduce adult attachment theory into LLM evaluation and use the Experiences in Close Relationships-Revised (ECR-R) scale to characterize attachment anxiety and avoidance. To evaluate emotional companionship capabilities of LLMs in realistic interaction scenarios, we present an emotional companionship benchmark, ECBench, spanning four scenarios including emotional support, collaborative tasks, conflict resolution, and social guidance, across friendship and romantic relationships. ECBench is utilized to assess model behavior using 11 dialogue-quality metrics and three evaluation methods. We evaluate the attachment tendencies of 32 LLMs and select representative models to investigate how these tendencies manifest in contextualized multi-turn interactions and whether they can be shaped through prompting. Our study provides a theoretical lens from psychology, along with practical tools to understand and select LLMs for emotional companionship.
How much phonological knowledge do language models (LMs) possess about Chinese characters, and how is this knowledge represented and used? We introduce PinyinBench, a controlled benchmark covering homophones, phonological features (initials, finals, and tones), and polyphonic characters. We investigate 17 text-only and multimodal LMs through three complementary approaches: behavioral evaluation, layer-wise probing, and activation steering. Behaviorally, models demonstrate substantial knowledge of initials and finals, with a frontier model (DS-v4-flash) scoring 90%+ in homophone judgment, when two characters share no phonological features. However, its accuracy drops sharply to 72% when two characters differ only in lexical tone. Probing reveals that tone is highly decodable from embeddings and improves modestly across layers, whereas segmental information starts less decodable and improves substantially with depth. Activation steering shows that segmental representations can systematically influence generation, while tone exhibits much weaker effects. Furthermore, audio-bearing training improves phonological decodability, including tone, without corresponding improvements in steering effectiveness. Together, these findings reveal distinct representational and functional profiles for segmental and tonal information, with tonal information much harder to access and manipulate, demonstrating the necessity for combined evaluation, probing, and causal intervention for understanding linguistic knowledge in LMs.
Adaptive sampling can substantially improve the value of sparse ocean observations by directing mobile robots toward measurements that are most informative for an evolving environment. Recent learned generative ocean models can produce high-resolution forecasts and ensembles at substantially lower computational cost than traditional numerical models, creating an opportunity to use rich predictive forecasts directly in closed-loop robotic sampling. However, these models do not natively provide the explicit uncertainty representations assumed by classical informative-planning methods, and their forecasts often evolve in learned latent spaces rather than directly in physical ocean variables. We develop a multi-robot closed-loop adaptive sampling framework that uses a single learned generative model for forecasting, uncertainty estimation, planning, and measurement assimilation. The model forecasts an ensemble in latent space, decodes it to physical ocean states, and uses ensemble statistics to construct A- or D-optimal sampling objectives. We select sampling locations subject to robot reachability under predicted currents. The collected measurements are assimilated back into the latent state using observation-conditioned diffusion sampling. Across 120 test scenarios A- and D-optimal sampling generally outperforms static, random, and systematic-coverage baselines while adding little planning overhead. Our experiments uncover several scenarios in which model uncertainty diverges from forecast error, motivating future work in uncertainty quantification.
ARLoop: Fully-Open Pre-trained Looped Transformers with Attention Residuals
Yuekun Yao, Yupei Du, Dawei Zhu, Huan Sun
AI 速览：提出 ARLoop——用注意力残差在模型深度上选择性聚合各层输入的循环 Transformer，并在 140M 至 1.2B 参数的真实预训练中验证其优于既有残差设计。
Looped transformers increase effective depth by repeatedly applying a shared set of transformer layers, offering a parameter-efficient way to scale computation. However, standard residual connections aggregate information across recurrent depth into a single stream, providing limited control over which intermediate representations from preceding layers should be reused. In this paper, we introduce ARLoop, a looped transformer architecture based on attention residuals, where the input of each layer is selectively aggregated through attention over model depth. Distinct from standard transformers, we find that looped transformers do not require access to the full residual history: sliding-window depth attention matches or outperforms full-history attention at larger recurrent depths. Across controlled experiments with multiple parameter scales (140M and 370M) and recurrent depths (R = 1 to 5), ARLoop consistently outperforms prior residual designs for looped transformers. More importantly, we further scale ARLoop to 0.6B and 1.2B parameters under a realistic LLM pretraining recipe, demonstrating that the gains transfer to realistic pretraining, with particularly strong improvements on math and coding tasks. Our 1.2B model, pretrained on only 0.5T tokens with three recurrences, outperforms Huginn-3.5B (the only fully-open looped LLM as of our release date) across all evaluated tasks and is competitive with similar-sized non-looped models such as Qwen3.5-2B despite being trained on substantially less data.
Large language model (LLM) agents can autonomously run neuroimaging analyses from research questions to reported findings. Existing agentic systems use tools to check that code runs and reported values match its output, but rely on LLM judgment to decide which results warrant scientific reporting. We show that LLM judges endorse site-confounded findings even when the confound is presented as evidence. We present \textbf{CONFIRM}, a framework that separates running an analysis from deciding which findings to report. CONFIRM encodes each hypothesis as a machine-readable, pre-specified analysis plan that fixes the claim and its pass criteria before computing results. It then filters results through five deterministic gates (search history and multiplicity, confounds, power, analytic stability, and replication) under three configurable reporting standards, from least to most demanding: \textit{Screen}, \textit{Replicate}, and \textit{Full}. Failed claims receive diagnoses that guide follow-up analyses. To evaluate reporting decisions at the claim level, we introduce \textbf{NeuroClaimBench}, comprising 489 executable neuroimaging claims with frozen data, including literature-supported findings and 150 controls designed to violate individual methodological requirements. On NeuroClaimBench, CONFIRM falsely supports at most 3\% of controls, compared with 23--38\% for the baselines, while recovering up to 61\% of literature-supported claims. When auditing an existing agent, CONFIRM reduces false support on controls from 23\% to 0\% while retaining the agent's literature-supported claims. Its diagnostic feedback also produces more supported revisions than Self-Refine with fewer LLM calls. Together, CONFIRM and NeuroClaimBench show that basing reporting decisions on pre-specified analysis plans and deterministic gates is essential for trustworthy autonomous scientific discovery.
World Action Models (WAMs) improve robotic manipulation by jointly modeling visual dynamics and robot actions. However, most existing WAMs model dense short-horizon visual dynamics, capturing local interactions but providing limited long-range task context. Extending prediction horizons increases uncertainty and computation while entangling task-level evolution with fine-grained local dynamics, obscuring their dependencies and distinct roles in action generation. We propose HiLo-WAM, which decouples long-horizon task evolution from local visual dynamics while jointly modeling their interaction with robot actions. A sparse long-horizon visual stream captures historical task context and future evolution, while a dense local dynamics visual stream models short-horizon robot-object interactions. The two streams operate at different temporal scales within a unified dual-DiT architecture and guide actions through structured cross-stream attention. Progressive inference further reduces unnecessary computation through stream-wise partial denoising, joint denoising, and dynamic skipping. Experiments demonstrate strong long-horizon manipulation performance: HiLo-WAM achieves a 93.70% average success rate on RoboTwin 2.0 and a 17.43 progress score on RoboDojo long-horizon tasks, with robust in-domain and OOD performance on the Fourier GR-3 robot. Ablations confirm the complementary contributions of long-horizon task evolution and local interaction dynamics, while temporal-scale analyses highlight the importance of balancing temporal coverage and resolution. Progressive inference further achieves a 2.93x speedup while improving performance.
Human speech includes nonverbal vocalizations (NVVs), such as laughter, sighs, breaths, and coughs, which convey affective and interactional information. Existing approaches typically represent NVVs as transcript-level tags, providing limited supervision for their waveform-time boundaries. We present NVV-Locator for fine-grained NVV temporal grounding. We first unify 26 NVV categories across public resources and construct large-scale timestamp-supervised training data through dual-LLM verification, transcript-guided forced alignment, and energy-based boundary refinement. We further introduce NVV-TimeBench, an expert-refined benchmark with 667 utterances and 1,094 events. NVV-Locator uses a non-autoregressive slot-filling architecture to jointly predict lexical timestamps, NVV categories, and event boundaries. On NVV-TimeBench, it achieves 71.0% Micro F1, 70.2% Macro F1, 80.4% Macro mIoU, and 59.6 ms Macro mMAE, outperforming the evaluated large audio model counterparts. Evaluation on an external corpus further demonstrates the cross-corpus generalization of NVV-Locator.
LLM recommender agents increasingly produce structured reports: sets of items accompanied by natural-language justifications. Yet existing benchmarks often focus on reranking small candidate sets or judging reports primarily by semantic plausibility, providing limited support for evaluating and training agents based on set-level utility. We introduce Recommender Atlas (Agentic Tasks for Learning and Assessment in Shopping), or RecoAtlas, a benchmark and reinforcement learning environment grounded in user behavior. RecoAtlas provides a shopping toolkit for product retrieval, complementarity, and diversity. Given a shopping query, an agent uses these tools to generate a structured recommendation report. Behavioral reference sets support exact-recovery evaluation and the training of a learned bundle-level reward that supplies feedback for policy optimization with RL. Its controlled tool environment exposes agents to semantic and behavior-aligned tools, enabling systematic analysis of how LLM reasoning, tool signals, and tool-use policies contribute to performance. Across controlled experiments, performance scales with model capacity and test-time compute, and improves with stronger tools, while semantic plausibility does not necessarily translate into behavior-grounded utility. RecoAtlas provides a foundation for training and evaluating shopping assistants to move beyond semantic plausibility toward coherent, behaviorally grounded recommendation sets.
Xuwei Tan, Jingchun Huang, Ming Cai, Shuangquan Lyu, Tiange Wang, Yinzhi Wang, Ronghua Shi, Jingwei Song 等 15 位作者
Multi-expert agent marketplaces -- where creators publish domain-specific AI experts for end users to consult -- are an emerging deployment pattern for language technology. A published expert can receive many requests yet little evidence about its creator, who remains active elsewhere: the cold-expert problem. We frame continuous cognitive distillation as converting attributed creator observations into revisable decision rules, communication patterns, and domain procedures. CogniLife proposes a cross-expert attribution design for this problem, with Creator Profiles for transferable creator patterns and Expert Journals for listing-specific knowledge; this design is not yet implemented or evaluated. A deployed platform partially realizes the surrounding lifecycle through creator-authorized authoring, capability-owned proposal filtering, and reversible consolidation of eligible owner-scoped memory. On ProfyBench-v1, an 85-task offline benchmark built from replay seeds and controlled internal tests, the Profy Full configuration as a whole achieves a mean overall score of 4.65/5 from the primary LLM judge, compared with 4.01 for Raw LLM and 4.28 for Expert Prompt, with the largest gains in completeness, specificity, and missing-information tasks. These configuration-level scores do not test cold-expert mitigation, cross-expert transfer, or longitudinal distillation.
Knowledge tracing is a key component of intelligent educational systems. However, most existing methods mainly focus on response prediction, providing limited support for explaining predicted outcomes and improving students' learning states. This paper proposes Diagnostic and Interpretable Knowledge Tracing (DIKT), which integrates response prediction, skill-level diagnosis, and model-based learning analysis. DIKT generates bounded cognitive-state trajectories and decomposes the final response probability into skill-wise probabilities. The trajectories reveal students' evolving strengths and weaknesses, while the skill-wise probabilities identify the skills that most strongly constrain the prediction. Based on these diagnostic results, DIKT supports direct cognitive-state intervention and effective-learning-duration-based what-if analysis to examine how hypothetical adjustments affect current predictions and subsequent state evolution. DIKT combines Bayesian multi-head attention, conditional variational recurrent inference, a logistic-growth decoder, and multiplicative skill-wise prediction. Experiments on three benchmark datasets show that DIKT outperforms all evaluated baselines, with an average absolute AUC gain of 2.04 percentage points. Other experiments further demonstrate that DIKT supports educationally meaningful diagnostic analysis and provides informative diagnostic indicators for downstream applications, extending knowledge tracing from response prediction to interpretable diagnosis and model-based learning analysis.
Man Liu, Xingchen Liu, Hairui Zhao, Guangming Tan, Dingwen Tao
The rapid scaling of large language models (LLMs) has made communication overhead a critical bottleneck in distributed model-parallel training. Although existing low-precision compression methods can effectively reduce communication costs, further reducing the communication footprint to ultra-low-bit formats such as FP4 introduces new challenges, including limited numerical range, sub-byte storage overhead, and increased computational cost from fine-grained quantization. In this paper, we present \textbf{FP4Quant}, a micro-grouped FP4 compression framework for efficient model-parallel LLM training. FP4Quant integrates a micro-grouped dual-scale FP4 quantization mechanism to preserve local tensor distribution characteristics, a warp-cooperative FP4 packing scheme to enable efficient sub-byte data representation, and a hybrid two-level FWHT pipeline to mitigate quantization errors with GPU-aware execution optimization. Experiments on NVIDIA H100 GPUs with Qwen-7B demonstrate up to 1.56$\times$ end-to-end throughput improvement over BF16, revealing a trade-off between throughput and model quality.
Reasoning-enabled model derivatives must turn intermediate deliberation into a usable answer within a deployment budget. We study that transition in an Ornith checkpoint adapted to multi-camera driving question answering with verifiable rewards. Paired archived outputs show that, on a selected 400-question development screen with a 256-token cap, GRPO increases answer extraction from 78.25% to 90.25% and accuracy from 55.50% to 63.75%. Of 39 wrong-to-correct transitions, 33 begin with an unextractable base response. The base's 87 unextractable responses all reach the token cap. Checkpoint-verified reconstruction on those same questions gives a much smaller first-pass accuracy gain at 512 tokens: +0.75 percentage points, versus +8.25 at 256. A later answer-cued continuation is analyzed separately; it helps the base more and narrows the adapter difference. On the full 2,511-question development set, this recovery-enabled pipeline reduces mean generated tokens from 108.6 to 47.1, while the paired accuracy interval includes zero gain. We frame these findings as adaptation of answer delivery under finite budgets, with explicit selection and provenance limits. Ornith's self-scaffolding lineage and newer reasoning-enabled derivatives such as Bonsai motivate the problem, but our measured results concern Ornith alone and do not establish a shared architectural mechanism or cross-model transfer.
A solid-state drive cannot overwrite data in place, so it cleans full blocks by copying out still-valid data, costing speed and lifetime. Interfaces such as NVMe Flexible Data Placement (FDP) let the host label data by expected lifetime so data dying together is stored together, and published schemes report cutting these copies by 20--75%. Those studies rarely state which blocks the drive chooses to clean. Varying only this choice in simulation, we show it decides how much a label is worth: label value peaks when the drive searches moderately for blocks holding little valid data, falls when it always takes the block with the least, and turns negative when it ignores how much valid data blocks hold. This holds in 47 of 48 synthetic configurations, and one real mail-server trace shows the same shape at about half the magnitude. Priced as spare capacity, a two-class label is worth 65% more spare when each scheme runs at its best setting, but becomes a net loss at 20% label error, typical of learned classifiers. Labels stating which data die together keep most of their value at 16 FDP handles and gain as cleaning improves. Evaluations must report the drive's cleaning policy.
Molecular docking predicts how a small molecule binds to a protein and is a key bottleneck in drug discovery. Classical scoring functions sum empirical pairwise contacts, blind to quantum-mechanical effects like orbital charge transfer that govern binding specificity. We introduce Q-Score, encoding GNN-predicted orbital donor-acceptor energies into a weighted graph and scoring binding by solving a maximum-weight vertex clique problem via Digitized-Counterdiabatic QAOA. Each interaction anchor maps to one qubit and compatibility constraints become edges. Across 11 protein targets, DC-QAOA recovers the exact optimum on 8 at 10 qubits. On 1000 AI-generated molecules, Q-Score is orthogonal to classical scoring with Spearman rho of 0.05, driven by orbital quality with rho of 0.90, and free of molecular-weight bias, enriching for strong orbital interactions at twice the random rate. DC-QAOA achieves a mean approximation ratio of 0.94 with 52 percent exact. Execution of 1000 circuits on IBM Eagle confirms 6-qubit solvability on NISQ hardware.
Wenqing Zheng, Dmitri Kalaev, Noah Fatsi, Daniel Barcklow, Owen Reinert, Igor Melnyk, Senthil Kumar, C. Bayan Bruss
Retrieval-Augmented Generation (RAG) systems rely critically on the retriever module to surface relevant context for large language models. Although numerous retrievers have recently been proposed, each built on different ranking principles such as lexical matching, dense embeddings, or graph citations, there remains a lack of systematic understanding of how these mechanisms differ and overlap. Existing benchmarks primarily compare entire RAG pipelines or rely on aggregate ranking metrics (Recall, MRR, nDCG) that assume item-level independence, providing little guidance on redundancy or complementarity between retrievers. This work presents MIGRASCOPE, a Mutual Information based RAG Retriever Analysis Scope. We introduce principled metrics grounded in information theory to quantify retrieval quality, redundancy/synergy, and marginal contribution. Although the framework is designed to be applicable to any setting where a ground-truth answer can construct a pseudo chunk distribution, the present empirical study focuses on multi-hop question answering - a demanding setting where retriever complementarity is especially pronounced. On four corpora and thirteen SOTA retrievers spanning lexical, dense, and graph-based families, we show that (i) the MI-based divergence metric captures retriever quality dimensions not reflected by Recall or MRR, (ii) graph-based retrievers are redundant with each other while BM25 remains informationally complementary, and (iii) in the multi-hop QA setting, a carefully chosen ensemble guided by MI-based attribution can outperform any single retriever. We position these findings as task-specific insights and provide a reusable framework for retriever analysis that practitioners can re-estimate on their own target domains.
To design platforms for multi-operator mobility markets coordinating their operations or infrastructure to support them, models are needed to understand the behavior of these operators and travelers in a shared network. Assignment games can prove useful, but operationalizing them requires adopting a more generalized aggregate assignment game framework that allows for identification and estimation. We extend such a framework to aggregate network assignment games, proposing a path-based formulation and a PURC-based formulation along with solution algorithms. An inverse optimization model is proposed for the link-based formulation. It is solved by a quadratic program that separates into a buyer block, which retains the node potentials and imposes every observed zero-flow cell, and a seller block. This estimator, introduced in the literature for uncapacitated route choice, is extended here to a capacitated two-sided market, with rationing prices supported on observed binding links and operator supply scales estimated by group. Computational tests with a toy network demonstrate the economic value of these models. A large-scale Long Island Rail Road network example (897 nodes, 671 census tracts, 123 LIRR stations, 8,320 directed links) is calibrated to 2026 fares, timetabled seat supply and National Transit Database operating costs. With 10-50% of links sampled, the misspecified inverse model fits oracle link flows with R^2>=96%. Held-out link error falls from 29% to 19%, correlations with the true rationing prices rise from 49% to 75%, and social optimal gross surplus is recovered within 6%. Deadweight loss is overestimated by 10-17%. Ablation tests show that ignoring the inactive cells more than doubles the held-out error and, at 30-50% sampling, roughly halves the correlation of the pricing wedge, while ignoring the rationing prices degrades every metric.
Sewoong Lee, Marc E. Canby, Ikhyun Cho, Julia Hockenmaier
The term "linear representation hypothesis" (LRH) has appeared across diverse subfields of artificial intelligence, neuroscience, and cognitive science. But previous works have not consistently treated the LRH as a falsifiable scientific hypothesis; we analyze these inconsistencies and examine their implications for how prior theoretical and methodological results should be interpreted. Based on this analysis, we argue that claims regarding linear representations become well-defined only through careful examination of the model, representation location, feature definition, and evaluation dataset. We therefore propose a more rigorous formalization of the LRH that makes these dependencies explicit and allows the hypothesis to be evaluated as a falsifiable scientific claim. Finally, we identify some non-trivial open problems that warrant further attention from the research community.
Jia Yu, Zilong Wang, Xinyang Jiang, Dongsheng Li, Shuo Wang
Computer-use agents could automate repetitive screen-based clinical work, but their reliability in medical graphical user interfaces remains largely unvalidated. Existing benchmarks focus on general web or desktop tasks and underrepresent medical software, which requires domain knowledge, exhibits markedly different UI design from mainstream applications, lacks public testing environments, and demands safety validation beyond task completion. We introduce MedCUA-Bench, an interactive benchmark for clinical computer-use agents. It covers 18 clinical scenarios across 10 medical domains, reconstructed from real product manuals and open-source medical systems to capture authentic clinical interfaces while avoiding licensing and privacy constraints. Each task ships with paired intent- and step-level goals to disentangle clinical reasoning from UI execution, and is evaluated by a deterministic checker over task completion and five clinical safety dimensions. Across 23 agents, the best closed-source model reaches 54.2% strict success, while all models remain below 9% on the real OpenEMR. Open-source agents average only 2.5%, with the best reaching 16.2%. MedCUA-Bench exposes the gap between current agents and reliable clinical software use, providing a reproducible testbed for future research.
By iteratively retrieving and integrating external evidence, multi-step search enables search-augmented agents to tackle complex question answering. However, vanilla GRPO assigns rewards based solely on final answers, providing limited guidance for intermediate search steps. To provide fine-grained process feedback, we evaluate the value of each search step: it should acquire evidence that advances problem solving while avoiding redundant retrieval. Building on these complementary criteria, we introduce BiCAA, which combines Forward Solvability Gain (FSG) and Subsequent Reasoning Gain (SRG) for bidirectional credit assignment. To reward progress in evidence acquisition, FSG quantifies changes in ground-truth answer likelihood after each interaction. SRG compares the subsequent action's likelihood with and without the current interaction to approximately measure its support for subsequent reasoning, using this support as a proxy signal for non-redundant search to encourage effective and efficient search. BiCAA calibrates local feedback through asymmetric fusion; for policy optimization, it integrates accumulated process rewards with terminal outcome supervision. Across seven open-domain QA benchmarks and two Qwen backbones, experiments show improvements in average F1, particularly on multi-hop tasks.
We present E-GenRM (E-commerce Generative Reward Model), a generative reward model tailored to e-commerce shopping agents. User expectations are shifting from simple product queries to open-ended shopping tasks. However, existing trajectories and reward designs remain largely bounded by current agent usage patterns and known failures, providing limited guidance for further capability improvement. These limitations call for an integrated reward modeling framework for open-ended shopping agent tasks. On the data side, E-GenRM leverages naturally occurring user behavior across the shopping app to construct diverse, long-horizon conversational trajectories beyond established usage patterns. On the modeling side, E-GenRM first learns from behavior-grounded evaluation chains through supervised fine-tuning. It is then optimized through criterion-guided reinforcement learning, which rewards correct preference judgments and evidence-supported factual claims under each criterion. Personalized hard-negative training further improves its sensitivity to user preferences reflected in shopping behavior, beyond explicit requirements. We additionally develop an active evidence retrieval extension that acquires missing e-commerce information for evaluation. On E-GenRM-Bench, E-GenRM-27B achieves 87.40\% preference accuracy, outperforming the strongest baseline by 6.97 percentage points. In a 14-day online A/B test on Taobao, a shopping agent post-trained with E-GenRM improves the effective shopping-assistance rate by 1.39 percentage points and user satisfaction by 2.33 percentage points.
Stanislav Kosorin, Trever Schirmer, Leonard Wossnig, David Bermbach
Workflow engines orchestrate pipelines of activities, which often call external APIs, on a fixed pool of worker slots, the threads that run them. When an API enforces a hard rate limit shared by all tenants and priorities, a throttled call that retries after a backoff keeps its worker slot and starves other tenants. Large Language Model provider APIs are particularly complex: their per-organization caps are low and bind independently on requests and tokens per minute. The operator can neither reconfigure the cap nor observe saturation before a request fails, so classical fair-queuing assumptions do not hold. We present Backpressure, a worker-side admission gate. It checks each call's cost atomically against shared admission state and enforces fairness over requests and tokens. A denied call releases its worker slot and waits on a durable timer, so waiting no longer blocks execution capacity. We compare against a naive retry, the AI gateway LiteLLM, and the built-in throttle of the workflow engine Temporal. On a workload with three priority classes, Backpressure raises fairness (Jain's index over each queue's fraction of its entitled share) from 0.60 to 0.89, where 1.0 is ideal. It also cuts high-priority tail latency roughly 8x. When tokens are the binding limit, all request-based baselines serve every priority equally, whereas Backpressure keeps the priority order and reaches a dominant-resource fairness of ~0.99. Admission costs one round trip to shared state, adds no measurable latency when the cap does not bind, and fails open if shared state is unavailable.