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

+1085新增论文
+884其中 AI 相关
+130已有论文修订
+59新增 AI 导读

arXiv 最近完成扫描:2026-11-18 04:28 · 全站最近入库:2026-11-18 19:45各信源进度 →

统计口径与数据新鲜度

统计 2026-11-17 21:42 — 2026-11-18 21:42(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:2026-11-18 04:28最近发现新增版本:2026-11-18 19:45库内论文最新更新时间:2026-11-18 08:00

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。

26,870 已收录论文22,374 AI 相关8,156 有修订112 HF 社区发现

浏览论文库

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 86 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-18 04:27(北京时间)
版本更新 · v3arXiv:2610.26825

MemT-1.0: Towards Memory-Native Evolution

Zhiyu Li, Zeyu Zhang, Linyun Liu, Yiwei Bu, Chenyang Xi, Xichong Zhang, Yi Yu, Yu Yu 等 74 位作者

原文摘要

Memory-native evolution requires making memory an intrinsic capability of foundation models, enabling them to continuously accumulate and utilize information. To this end, we introduce MemT-1.0, a new architecture of memory foundation models that internalize memory as part of model computation. Building on Metis, MemT-1.0 employs Gated DeltaNet-2 with a stable token-level memory update mechanism within each turn. We further introduce Mixture-of-Storage to scale native memory capacity. Following a deliberate backbone preparation, we develop a two-stage training pipeline consisting of continued pre-training and post-training, supported by 34B tokens of training data and efficient training and inference infrastructure. We release a model family comprising MemT-1.0-Nano, MemT-1.0-Flash, and MemT-1.0-Pro, and conduct a comprehensive evaluation of memory and general capabilities. The results show that they outperform the parametric memory baselines on average and remain competitive with retrieval-augmented methods, while largely preserving general capabilities. At a history length of 128K tokens, MemT-1.0-Nano achieves approximately $1.86\times$ end-to-end inference speedup and a $66\times$ reduction in persistent storage relative to full context KV cache. By internalizing retrieved procedural experience, MemT-1.0 improves task success on agentic tasks, providing a practical step toward memory-native evolution.

cs.CL · cs.LG

新论文 · v1arXiv:2611.22778

A Full-Ledger Counterfactual Audit of Adaptive Reasoning

Xiaopeng Chu, Wenzheng Kong, Wei Liu, Wenyi Zhang

原文摘要

Adaptive reasoning must decide whether to answer now or spend more compute. Predicting that the current answer is wrong does not establish whether further reasoning will improve it. We introduce the Full-Ledger Counterfactual Value Audit (FLCVA), which compares an immediate answer with four fixed-length continuations from the same frozen prefix under one answer procedure and accounts for the cost of every signal used by a policy. On GSM8K and MATH-500, a random 256-token continuation recovers a wrong answer in 26.6% of Qwen3-8B cases and 18.9% of DeepSeek-R1-Distill-Qwen-7B cases, while damaging a correct answer in 1.4% and 1.9%. Correctness labels from independently sampled continuations predict a separate continuation's outcome in both primary models. In offline replay, using these labels to allocate compute improves performance on DeepSeek; the Qwen3 gain remains inconclusive. Under full cost accounting, the tested lightweight, myopic controllers mainly recognize current errors and add little continuation-specific value beyond that shortcut. A Llama-8B replication and action/cost ablations examine the robustness of these findings. FLCVA separates measuring local continuation value, predicting it from observable signals, and using those predictions to allocate computation.

cs.AI · cs.CL

版本更新 · v3arXiv:2607.28591

Change2Task: From Repository Changes to Executable Coding Agent Tasks and Environments

Haomin Qi, Xingliang Wang, Xuanqi Gao, Baihui Sang, Xin Zhang, Minghua Ma, Pengfei Gao, Yu Kang 等 12 位作者

原文摘要

Coding agents learn from and are measured on executable tasks, but building such tasks from repository history usually means rebuilding an environment at an old revision for every task, which is costly and brittle. We observe that a merged pull request (PR) can instead be undone on today's code: removing the effect of a bug fix or feature from a maintained revision recreates the same task inside an environment that already builds and runs. We present Change2Task, which converts merged PRs into verified tasks on an up-to-date revision of the same repository. It reconstructs the pre-change state by patch reversal, code mapping, or agent-based rewriting, and keeps a task only if its target tests pass, fail, and pass again across the healthy, task, and restored states. From 1,130 PRs spanning bug fixing, feature addition, test generation, API migration, and security repair, Change2Task builds 900 verified tasks, recovering 29% more than SWE-smith's PR Mirror on matched bug-fix candidates. The reconstructed tasks preserve the evaluation signal: in aggregate, four coding agents solve 41.1% of tasks on both the original and reconstructed versions, with 89.7% per-task agreement. Sharing modern environments across tasks further reduces setup time by 58% and storage by 71%. On 11 private industrial repositories, Change2Task yields 1,000 tasks on which the strongest model reaches only 14.6% Pass@3.

cs.SE · cs.CL · cs.LG

版本更新 · v4arXiv:2604.07789

ORACLE-SWE: Quantifying the Contribution of Critical Information Signals on SWE Agents

Kenan Li, Qirui Jin, Liao Zhu, Xiaosong Huang, Yijia Wu, Yikai Zhang, Xin Zhang, Zijian Jin 等 16 位作者

原文摘要

Recent advances in Language Model (LM) agents have increasingly focused on repository-level software engineering (SWE) tasks, leading to various agentic workflows, training methods and failure analysis. Interestingly, rather than directly generating or evaluating edit patches, these approaches first examine and improve intermediate information signals collected during agent exploration, including Reproduction Test, Regression Test, Edit Location, Execution Context, and API Usage. Although different studies examine different subsets of these signals, no work has systematically compared their relative effectiveness in improving final task success or measured the remaining space for improvement in each signal. To address this gap, this work quantifies the contributions of both oracle and agent-extracted signals to final task success and measures the gap between current agent-found signals and their oracle counterparts. We hope our findings could suggest future research emphasis in SWE agents. We release our source code and oracle-signal dataset as a reusable methodology to benefit future signal analysis of SWE agents.

cs.MA · cs.CL · cs.SE

版本更新 · v3arXiv:2510.08697

BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution

Terry Yue Zhuo, Xiaolong Jin, Hange Liu, Juyong Jiang, Tianyang Liu, Chen Gong, Bhupesh Bishnoi, Vaisakhi Mishra 等 40 位作者

原文摘要

Crowdsourced model evaluation platforms, such as Chatbot Arena, enable real-time evaluation from human perspectives to assess the quality of model responses. In the coding domain, manually examining the quality of LLM-generated content is extremely challenging, as it requires understanding long chunks of raw code and deliberately simulating code execution. To this end, we introduce BigCodeArena, an open human evaluation platform for code generation backed by a comprehensive and on-the-fly execution environment. Built on top of Chatbot Arena, BigCodeArena enables the execution of LLM-generated code and allows humans to interact with the execution process and outcomes. We collected over 14,000 raw code-centric conversation sessions across 10 widely used LLMs, spanning 10 languages and 8 types of execution environments. Among these conversations, we identified more than 4,700 multi-turn samples with pairwise human preferences. Further analysis uncovers underexplored preferences of LLMs in fine-grained domains characterized by tasks, languages, and frameworks. To systematically examine code understanding and generation capabilities of frontier LLMs, we curated two benchmarks based on the collected data, namely BigCodeReward and AutoCodeArena. For BigCodeReward, we post-processed the 4,700 conversations and evaluated the consistency between reward models and human preferences. The evaluation shows that most LLMs have superior performance in judging coding preferences when the execution results are available. Inspired by these findings, we propose AutoCodeArena, an automatic Elo rating benchmark designed to assess the coding quality of LLMs without human involvement. We find that proprietary LLMs like GPT-5, Claude-Sonnet-4, and Claude-Opus-4 still lead in code generation performance among recent emerging models.

cs.SE · cs.AI · cs.CL

版本更新 · v5arXiv:2406.15877

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

Terry Yue Zhuo, Minh Chien Vu, Jenny Chim, Han Hu, Wenhao Yu, Ratnadira Widyasari, Imam Nur Bani Yusuf, Haolan Zhan 等 33 位作者

原文摘要

Task automation has been greatly empowered by the recent advances in Large Language Models (LLMs) via Python code, where the tasks ranging from software engineering development to general-purpose reasoning. While current benchmarks have shown that LLMs can solve tasks using programs like human developers, the majority of their evaluations are limited to short and self-contained algorithmic tasks or standalone function calls. Solving challenging and practical tasks requires the capability of utilizing diverse function calls as tools to efficiently implement functionalities like data analysis and web development. In addition, using multiple tools to solve a task needs compositional reasoning by accurately understanding complex instructions. Fulfilling both of these characteristics can pose a great challenge for LLMs.To assess how well LLMs can solve challenging and practical tasks via programs, we introduce BigCodeBench, a benchmark that challenges LLMs to invoke multiple function calls as tools from 139 libraries and 7 domains for 1,140 fine-grained tasks. To evaluate LLMs rigorously, each task encompasses 5.6 test cases with an average branch coverage of 99%. In addition, we propose a natural-language-oriented variant of BigCodeBench, BigCodeBench-Instruct, that automatically transforms the original docstrings into short instructions only with essential information. Our extensive evaluation of 60 LLMs shows that LLMs are not yet capable of following complex instructions to use function calls precisely, with scores up to 60%, significantly lower than the human performance of 97%. The results underscore the need for further advancements in this area.

cs.SE · cs.AI · cs.CL

版本更新 · v2arXiv:2610.38610

为谁修复?选择失败LLM智能体修复方案时的评估者-执行者错配

Repairs for Whom? Evaluator-Executor Mismatch in Selecting Repairs for Failed LLM Agents

Yijun Liao, Fanwei Liang

AI 速读 基于摘要

作者报告:在固定失败轨迹与候选单步修复的条件下,用实际部署策略的独立rollout来评估修复选择,比贪心回放、更强模型或微调后继模型更贴近部署价值。

cs.SE · cs.CL

查看原文摘要

When an LLM agent fails, repair methods choose which action to replace and what to put in its place. They often score candidate repairs with greedy replay, a stronger model, an environment oracle, or a judge that reads the failed log. We ask whether these evaluators pick repairs that help the agent that will actually execute them. Holding failed trajectories and candidate single-step repairs fixed, we label every candidate with independent rollouts of the deployed policy and measure how much deployment value each evaluator's choice loses. On failures not screened for repair value, pre-specified tests show that greedy replay loses 0.08-0.13 in expected net success gain in ALFWorld and 0.06-0.13 in net reward in WebShop, relative to selection with the deployed agent's own rollouts. At an equal rollout budget, selecting repairs for Qwen2.5-7B with Qwen2.5-14B rollouts loses 0.25 in ALFWorld and 0.07 in WebShop, twenty 14B rollouts per candidate are worth no more than two of the 7B agent's own, and fresh rollouts confirm that the two models prefer opposite repairs on selected disagreement cases. The agent's own fine-tuned successor is a worse evaluator still: after one round of training on verified repairs, its rollouts lose 0.55-0.60 when they select repairs for the agent it was trained from, as a pre-specified test confirms. Relative to what selection is worth, substitute models forgo 11-53%; on $τ^2$-bench, where selecting is worth only 0.10, they lose 0.03. Repairs should be verified with the executor that will run them, and repair labels should name their executor, including its training version.

新论文 · v1arXiv:2611.16384

Beyond Compressing Everything: Contrastive Memory Learning for Selective Context Compression

Sikui Zhang, Haowei Liu, Ziyun Gan, Chunfeng Yuan, Guangze Gao, Nian Liu, Boyu Cai, Bo Li 等 12 位作者

原文摘要

As large language models (LLMs) advance toward increasingly long inputs, soft context compression offers a promising approach to efficient inference by converting long inputs into compact memory slots. Under limited memory budgets, effective context compression remains challenging because task-relevant evidence varies substantially across queries, ranging from localized evidence to multiple distant spans or information dispersed throughout the full document. Instead of compressing everything indiscriminately, we propose CoMem, a query-guided selective context compression framework that learns memory representations conditioned on both the document and the query. CoMem develops Contrastive Compressor Learning to achieve query-sensitive yet context-preserving memory representations. To inject a strong prior over query-dependent evidence distributions, we devise a same-context hard negative mining strategy by pairing a single document with multiple queries, requiring the compressor to selectively retain information according to each query's distinct evidence needs. To address the restricted information flow imposed by causal masking, we further enable bidirectional attention among memory slots. Experiments demonstrate that CoMem consistently outperforms existing context compression methods across compression ratios on question answering and RULER-NIAH benchmarks. Our code is available at https://github.com/scar-on/CoMem.

cs.CL · cs.AI

版本更新 · v5arXiv:2508.02091

CRINN: Contrastive Reinforcement Learning for Approximate Nearest Neighbor Search

Xiaoya Li, Albert Wang, Guoyin Wang, Chris Shum, Jiwei Li

原文摘要

Approximate nearest-neighbor search (ANNS) algorithms have become increasingly critical for recent AI applications, particularly in retrieval-augmented generation (RAG) and agent-based LLM applications. In this paper, we present CRINN, a new paradigm for ANNS algorithms. CRINN treats ANNS optimization as a reinforcement learning problem where execution speed serves as the reward signal. This approach enables the automatic generation of progressively faster ANNS implementations while maintaining accuracy constraints. Our experimental evaluation demonstrates CRINN's effectiveness across six widely-used NNS benchmark datasets. When compared against state-of-the-art open-source ANNS algorithms, CRINN achieves best performance on three of them (GIST-960-Euclidean, MNIST-784-Euclidean, and GloVe-25-angular), and tied for first place on two of them (SIFT-128-Euclidean and GloVe-25-angular). The implications of CRINN's success reach well beyond ANNS optimization: It validates that LLMs augmented with reinforcement learning can function as an effective tool for automating sophisticated algorithmic optimizations that demand specialized knowledge and labor-intensive manual refinement. Code can be found at https://github.com/ornith-ai/CRINN

cs.LG · cs.AI · cs.CL · cs.DB

新论文 · v1arXiv:2611.14908

Hy ASR 3.0 Technical Report

Aoxiong Yin, Auden Li, Chao Huang, Chuanyang Hu, Eitan Wang, Ethan Zou, Fangming Feng, Guangyu Chen 等 63 位作者

原文摘要

We present Hy ASR 3.0, a large language model (LLM)-based automatic speech recognition (ASR) system that delivers strong recognition accuracy, supports flexible contextual conditioning, and maintains robust recognition performance across diverse speech conditions. Hy ASR 3.0 supports Mandarin Chinese, English and 13 Chinese dialect groups, including Cantonese. Among the evaluated systems, it achieves the lowest aggregate error rates on public Mandarin Chinese, English, and Cantonese benchmarks and the lowest error rates on the corresponding internal test sets, alongside first-tier dialect recognition performance. Beyond audio-only transcription, Hy ASR 3.0 supports multiple forms of external context, including hotword lists, question--answer dialogues, and summaries, enabling the use of task-relevant information during recognition. The model also leverages intra-utterance context to disambiguate homophonic and phonetically similar keywords using information from other parts of the same utterance, without requiring a separate contextual input. Evaluations demonstrate strong contextual recognition in both external-context and intra-utterance settings, as well as robust recognition of long-form speech, children's and older adults' speech, low-volume speech, and speech in noisy environments. This report describes the model architecture and training methodology and presents evaluations across recognition accuracy, context utilization, and robustness.

cs.CL · eess.AS

新论文 · v1arXiv:2611.14861

Is Jev a Good Multilingual Judge?

Wei Liu, Yi Fan, Xiyan Fu

原文摘要

Generative large language models are widely used for multilingual text evaluation, yet whether decision models that directly output judgment probabilities can serve as multilingual judges has received limited attention. We evaluate Jev against generative LLM judges on four multilingual benchmarks with human annotations, covering up to 47 language varieties and three evaluation formats: binary decisions, pairwise comparisons, and graded ratings. At approximately $1/31$ to $1/61$ of GPT-6.1's evaluation cost, Jev is competitive on most multilingual judging tasks, but falls behind GPT-6.1 on knowledge-intensive answer verification. Further analysis shows that Jev is more likely to accept incorrect answers in low-resource languages than in English and that its probability outputs help identify errors across tasks. Based on these findings, we offer recommendations on evaluation instructions, confidence-based selection, and model routing, providing empirical guidance for using Jev as a multilingual judge.

cs.CL

版本更新 · v2arXiv:2610.37105

Text Generation in Low-Resource Languages Using Human-Curated Vocabularies

Michael Christlambert Sinanta, Muhammad Dehan Al Kautsar, Amanda Nurul Izzah, Thanh-Nhi Nguyen, Phan Phuong Mai Chau, Zheng Wei Lim, Fajri Koto

原文摘要

Synthetic text generation offers a promising approach to alleviating data scarcity in low-resource languages. However, its effectiveness is inherently limited for languages that are scarcely represented, or entirely absent, in an LLM's pretraining data. At the same time, many such languages have human-curated vocabularies that provide rich lexical and semantic information, often accompanied by grammatical information such as part-of-speech labels. In this work, we ask whether LLMs can leverage these resources to generate coherent text in extremely low-resource and unseen languages. We introduce a framework that combines human-curated vocabularies with a small number of human-written exemplars that provide syntactic signals, while sampling candidate words according to a Zipfian distribution. Across languages with varying levels of resource availability, we evaluate the generated text using lexical diversity, off-target ratio, and human judgments of fluency. Our results show that reasoning-enabled LLMs can generate fluent text even in extremely low-resource languages, with particularly strong results at the sentence level and promising performance at the paragraph level. Our findings highlight a practical pathway for bootstrapping synthetic data in low-resource languages, expanding the amount of usable text even when existing digital resources are extremely limited. Our source code is available at https://github.com/michaelsinanta/vocab2text.

cs.CL

版本更新 · v2arXiv:2611.10693

Machine Hearing: Can Audio LLMs Generalize Beyond Human Auditory Perception?

Liwei Liu, Jun Tao, Xingyuan Li, Xuenan Xu, Chen Zhang, Junyi Wang, Boya Zhang, Xiaoyu Yang 等 11 位作者

原文摘要

Multimodal LLMs have made remarkable progress in perceiving the physical world, while their perceptual scope remains largely human-centric, focusing on what humans can directly see or hear. This progress raises the prospect of perception beyond human limits, such as extending hearing to sounds that are inaudible to humans. More broadly, the physical world is full of waves, vibrations, and oscillations, only a small fraction of which are directly audible to humans. This invites the extension of multimodal LLMs toward a broader form of machine hearing, which remains largely underexplored. To investigate this direction, we introduce the Machine Hearing Benchmark (MHB), comprising 6,128 instances across 13 tasks and seven domains. MHB spans highly heterogeneous input regimes, with sampling rates from approximately $10^{-5}$ to $10^6$ Hz, sequence lengths from $10^1$ to $10^6$ samples, and both single- and multi-channel inputs. Evaluations of current open-source and proprietary Audio LLMs (ALLMs) reveal limited zero-shot transfer to machine hearing. Supervised adaptation on machine-hearing data alone does not close this gap, suggesting that the challenge extends beyond data exposure to a systematic mismatch between existing audio front ends and machine-hearing inputs. To address this mismatch, we introduce MachEar, a lightweight encoder extension that complements the pretrained Audio LLM for machine-hearing inputs. The results show that MachEar equips a pretrained Audio LLM with stronger machine-hearing capabilities while largely preserving its existing audio understanding performance.

cs.SD · cs.CL

版本更新 · v5arXiv:2601.21343

Self-Improving Training & Thinking

Ellen Xiaoqing Tan, Jack Lanchantin, Zichun Yu, Polina Kirichenko, Shehzaad Dhuliawala, Danwei Li, Thao Nguyen, Jing Xu 等 14 位作者

原文摘要

Large language models are classically trained in stages: pretraining on raw text followed by post-training for instruction following and reasoning. However, this separation creates a fundamental limitation: many desirable behaviors such as safety, factuality, overall generation quality, and reasoning ability are only added at a late stage, even though the patterns learned earlier strongly shape a model's capabilities. To tackle this issue, we introduce a new way to pretrain and mid-train models that incorporates these behaviors earlier. We utilize an existing strong, post-trained model to both rewrite data and to judge policy model rollouts, thus introducing thinking and reinforcement earlier in training. We further improve reasoning at all stages by introducing the concept of self-improving post-thinking: improving the model's reasoning and self-judging ability by reasoning after generation. In our experiments, we show these approaches can give strong gains across multiple axes.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.11487

THEATRON: A Benchmark for Evaluating Persona Persistence in Long-Horizon LLM Decision-Making

Yuyan Wang, Yuechen Jiang, Zhiwei Liu, Jimin Huang, Sophia Ananiadou

原文摘要

Existing evaluations often test persona expression rather than whether persona information persists in extended decision-making. We introduce THEATRON, a controlled benchmark that uses financial replay to evaluate persona persistence in long-horizon sequential decision-making. In these environments, models conditioned on fixed behavioral-finance personas observe shared daily market evidence and trajectory-specific account-state feedback and make sequential decisions over multi-month trajectories. We evaluate Zero-Shot, One-Shot, and Chain-of-Thought persona injection across four LLMs, using cumulative-return and action trajectories to diagnose within-persona stability, cross-persona separability, and action collapse. Persona conditioning produces systematic behavioral shifts beyond a no-persona baseline, and persona-level separation generally exceeds the variation induced by semantically equivalent descriptions. Nevertheless, these shifts do not reliably form coherent persona-specific policies: apparently stable separation can arise from stereotyped action collapse, while diverse policies can exhibit noisy trajectory switching. These findings expose a gap between observable persona-induced behavior and persistent, context-sensitive persona-conditioned decision policies. Code and benchmark data are publicly available at: https://github.com/yuyan12138/theatron

cs.CL

版本更新 · v2arXiv:2602.21215

Inference-time Alignment via Sparse Junction Steering

Runyi Hu, Jie Zhang, Shiqian Zhao, Jiale Meng, Jiwei Li, Jason Zeng, Ming Wu, Michael Heinrich 等 10 位作者

原文摘要

Token-level steering has emerged as a pivotal approach for inference-time alignment, enabling fine-grained control over large language models (LLMs) by modulating their output distributions without parameter updates. While effective, existing methods rely on dense intervention at every decoding step. This persistent manipulation not only incurs substantial computational overhead but also risks compromising generation quality by excessively drifting from the model's intrinsic distribution. In this work, we show that dense intervention is unnecessary and propose Sparse Inference-time Alignment (SIA), which performs sparse junction steering, intervening only at critical decision points along the generation trajectory. Our key insight is that high-entropy junctions are empirically consistent with pivotal decision points in the generation trajectory and are particularly susceptible to misalignment, suggesting that these points benefit from alignment-related reward signal. Extensive experiments across different model families and alignment objectives show that steering only 20%-80% of tokens achieves the superior alignment-efficiency trade-offs. For strong base models such as Qwen3, intervening on as few as 20% of tokens matches or even surpasses heavily post-trained instruct models. This sparsity enables stronger guidance while better preserving the model's native distribution, integrates seamlessly with search-based methods (e.g., Best-of-N), and improves efficiency by reducing the required sampling budget while yielding practical wall-clock speedups under matched-quality comparisons.

cs.CL · cs.AI

新论文 · v1arXiv:2611.09491

An Effective Spoken Language Understanding Approach Leveraging Query-Adaptive Intent Detection

Bing-Long Kuo, Yu-Hsuan Fang, Tien-Hong Lo, Hsueh-Wei Liao, Berlin Chen

原文摘要

Large language models enable fine-tuning-free spoken language understanding (SLU) through in-context learning, but direct query-based retrieval may select semantically similar yet intent irrelevant demonstrations. This paper presents Query-Adaptive Intent Detection via Prototype Extraction (APEX), a two-stage retrieval framework that first selects candidate intents using intent-level prototypes and then retrieves contextual exemplars within the selected intent space. Experiments on the SLURP dataset evaluate joint SLU performance using Exact Match Accuracy (EMA) under full-transcript and perplexity-stratified conditions, with additional retrieval analysis on ASR-generated transcripts. Compared with query-based retrieval, APEX improves EMA by 4.34 and 3.97 percentage points with Qwen3 4B and Gemma-3-4B, respectively, on full transcripts. Under high-perplexity inputs, APEX also yields EMA gains of 7.66 and 2.13 percentage points. Retrieval analysis further indicates that prototype-guided candidate selection achieves higher ground truth intent coverage across multiple input conditions.

cs.CL

新论文 · v1arXiv:2611.07605

The Gating Mechanism of KV Cache Eviction

Hao Gu, Xintong Yang, Zhiwei Li, Lei Zhu, Jian Yang, Qiyuan Zhu, Dezhi Li, Jingyuan Zhang 等 13 位作者

原文摘要

KV cache eviction must decide which entries carry information that the rest of the cache can no longer supply. A natural way to estimate this is a learnable gate: give each cached entry a gate on its attention weight and train the gate on the model's own objective. The gate belongs inside the normalizing softmax, where suppressing one entry hands its mass to the others rather than discarding it. That same normalizer absorbs any per-query constant, so a constant gate cancels and reproduces the frozen model exactly. Trained on the language-modeling loss the gate stays near a constant score while its loss falls throughout: optimization succeeds, but a frozen backbone was fit for dense attention, so its loss prefers the no-op, and a near-constant score ranks nothing. The no-op is not a floor the gate improves from but what its objective asks for. A second obstacle survives once the gate moves, since the same gradient charges the gate for errors the frozen model already makes, which no reweighting of positions separates out. Much of the machinery in learnable eviction, from sparsity penalties to relaxations of the discrete decision, works around the first obstacle rather than removing it; both are removed by putting dense attention beyond the gate's reach and making the dense model's predictions its target. Under these two conditions capacity is never the missing ingredient: neither a larger scorer, nor explicit access to the past, nor a trainable backbone helps. The same holds for what eviction unavoidably discards, which is recovered by parameter-free centroids read inside the same softmax far better than by any memory read outside it, since the mass they must return is a share of that normalizer and available nowhere else. At 75\% KV cache compression, this gate reaches 90.14 on RULER at 8K and 46.03 on LongBench with Qwen3-8B, 3.72 and 1.36 points below full-cache inference.

cs.CL

新论文 · v1arXiv:2611.06976

On the Opportunities and Challenges of Prefill-to-Autoregressive Cross-Model Transfer

Zhenyi Shen, Junru Lu, Wei Liu, Jiangnan Ye, Jiazheng Li, Di Yin, Xing Sun, Tameem Adel 等 11 位作者

原文摘要

Prefill-to-Autoregressive Cross-model Transfer (PACT), a form of latent communication between language models, combines a larger model's prompt representations with a smaller model's efficient generation, offering a promising way to balance capability and inference cost. However, existing systems provide limited support for independently evaluating large-small model collaboration, and improved task performance alone does not establish whether the receiver benefits from the sender's information. We introduce PLatour, an open-source PACT framework that connects frozen models through lightweight, zero-initialized linear bridges trained directly with task rewards and supports hybrid-attention architectures. PLatour significantly outperforms open-source baselines on generation tasks and surpasses Mostik's reported results on shared mathematics and coding benchmarks using a substantially smaller sender. However, controlled experiments show that the gains on some tasks can be reproduced by uninformative sender signals, self-communication, or receiver-only training. We find that trainable bridges can act as parameter-efficient adaptation mechanisms without exploiting useful sender information. We therefore propose PACT-Check to quantify how much of a trained PACT system's gain genuinely comes from communication. Finally, we discuss the opportunities and challenges of PACT systems. Code is available at https://github.com/zhenyi4/platour.

cs.CL

新论文 · v1arXiv:2611.06185

Editing Large Language Models with Geometry-Aware Regularization

Bingqing Liu, Wei Liu, Jun Wang, Xiaobo Sun, Yuhua Li, Ruixuan Li

原文摘要

Knowledge editing has become a key technique for updating knowledge in large language models. However, existing methods neglect the geometric structure of the knowledge representation space, leading to two key limitations: they cannot distinguish which pre-trained knowledge should be preferentially preserved (core vs. non-core) nor which new edits are inherently difficult (hard vs. easy). To address this, we propose GEdit, a geometry-aware framework built on a Low-Rank Plus Diagonal (LR+D) factor model that captures the intrinsic low-rank structure of Transformer FFN hidden representations. GEdit features two complementary mechanisms. First, the low-rank regularization applies targeted protection on a low-rank core knowledge subspace. Second, the adaptive regularization dynamically adjusts the regularization strength based on each edit request's editability: relaxing constraints for hard edits to ensure success, and tightening them for easy edits to better preserve pretrained knowledge. Extensive experiments on GPT2-XL, GPT-J, and LLaMA-3-8B under large-scale editing scenarios, including both mini-batch and fully sequential editing settings with up to 10,000 continuous updates, show that GEdit consistently and significantly outperforms state-of-the-art baselines in both editing success and knowledge preservation.

cs.CL · cs.AI

版本更新 · v2arXiv:2608.02602

AURORA-LM: Autoencoding Unified Representation for Continuous-Latent Diffusion Language Modeling

Jiajun Liang, Yucheng Liao, Yukang Cao, Jiazhe Wei, Ken Li, Wende Tan, Jiankun Zhang, ZY Cui 等 15 位作者

原文摘要

Language remains an outlier in generative modeling: while images, video, and audio are increasingly modeled in continuous latent spaces, text generation still relies predominantly on discrete tokens. Existing continuous language models either inherit embedding spaces not designed for joint generation and decoding, or compress autoencoded latents to ease diffusion, sacrificing token-level fidelity. Instead of simplifying the representation to suit the generative model, we preserve a high-capacity, decodable text latent and design the diffusion model to learn its distribution directly. We introduce AURORA-LM, a continuous-latent diffusion language model that separates the construction of a decodable text representation from the modeling of its distribution. A Query-based Encoder-Decoder organizes text into a high-capacity, prefix-aligned latent sequence, and a Block-causal Diffusion Transformer learns its distribution through flow matching, generating blocks left to right while denoising positions within each block in parallel. Because such a latent is harder for diffusion to model, AURORA-LM restricts only the noisy-input pathway while retaining the full clean-latent prediction target, accommodating full-width latents without reducing decoder-facing capacity. We further calibrate the noise-level distribution to the latent width, and introduce self-trajectory consistency to bridge independently sampled training noise and iterative denoising at inference. AURORA-LM achieves the strongest performance among evaluated continuous and diffusion-based language models on OpenWebText free generation and XSum summarization. Scaling to 1B parameters with about 1500 EFLOPs of total compute yields further gains, surpassing a larger publicly released latent-diffusion language model under a matched evaluation protocol. All experiments are conducted on Ascend NPUs.

cs.CL

新论文 · v1arXiv:2611.03992

FlagRot: Selective Rotation for Low-Bit KV Caches

Kunming Shao, Xijie Huang, Jierun Chen, Jiaqi Shao, Yanli Wang, Jiangnan Yu, Qiwei Li, Wei Zhang 等 10 位作者

原文摘要

Long-context inference is bound by the KV cache, which 2-bit codes could shrink about sixfold if their error stays in check. The usual remedy rotates keys before quantizing, assuming rotation always helps. However, we observe that it does not: rotation flattens a spiky piece of a key but can give a flat one a new peak, and both coexist within one key. We therefore propose FlagRot, which stores each 32-channel piece rotated or as is, whichever has the smaller calibrated attention error, and rotates the query instead when reading. We further show that how often to decide follows the quantization axis: per-channel caches such as KIVI need one decision per head, at no write cost, while per-token caches decide per key, with the choice stored in existing sign bits. At 2.5 bits per value, FlagRot lowers the twelve-model KL divergence of KIVI from 18.7 to 12.4 x 10^-3, below RotateKV on 9 of 12 models. Served as a vLLM plugin, it raises RULER at 128k tokens by +2.9 points and the LongBench v1 mean by +0.4 to +1.9 points on the three of four models where 2-bit KIVI loses accuracy. It also lowers the error of 4-bit integer and FP4 caches.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.03634

WebHarbor: Docking Real Websites for Evolving GUI Agent Environments

Zhaoyang Wang, Qianhui Wu, Boyu Gou, Demi Wang, Jiacheng Zhang, Peichun Hua, Qize Li, Shi Qiu 等 57 位作者

原文摘要

GUI web agents require realistic and interactive environments that remain stable during both evaluation and training. Live website evaluation suffers from changing content, access restrictions, and limited coverage of authenticated workflows, while existing synthetic environments for training often lack functional depth and visual fidelity. We propose WebHarbor, a suite of 153 executable and dockerized website environments that faithfully reconstruct the visual interfaces, content, and functional workflows of real-world web services under deterministic control. We build WebHarbor through task-driven human-agent collaboration, where environments evolve as new tasks expand their functional scope, coding agents implement the required changes, and human contributors audit visual fidelity, functional correctness, and task validity. The resulting suite comprises 3,518 tasks spanning multi-step and post-authentication workflows, where existing works often fail to cover. Beyond reliable evaluation, WebHarbor provides high-fidelity synthetic web environments for cost-efficient reinforcement learning (RL). We demonstrate that a Qwen 4B-level GUI agent trained with RL in WebHarbor generalizes to unseen tasks and performs competitively with leading open-source GUI web agents across four benchmarks. The code and data are publicly available at https://github.com/aiming-lab/WebHarbor.

cs.AI · cs.CL · cs.LG

新论文 · v1arXiv:2611.02079

ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost

Xingzhi Qian, Sungho Park, Rongyuan Tan, Jianbo Yu, Jue Zhang, Lorenzo Cavallaro, Menglin Xia, Yanzhi Chen 等 16 位作者

原文摘要

Harness optimization can substantially improve LLM agents without modifying model weights, but existing methods primarily optimize response quality. Extending them to joint quality--cost optimization introduces a challenge: applying the deployment preference during intermediate updates can prematurely reject candidates that are valuable for later improvements. We introduce ParetoSaddler, which addresses this mismatch by separating intermediate candidate promotion from final deployment selection. At each iteration, an optimizer updates a parent harness to produce a candidate, and then compares the two over repeated executions and uses Bayesian bootstrap Pareto support to promote candidates showing credible progress in either quality or cost. Promoted candidates receive broader development evaluation and remain available as parents for later optimization steps. We evaluate ParetoSaddler on ClawMark and GAIA2 using domain-disjoint splits. On ClawMark, ParetoSaddler improves quality from 70.86\% to 81.35\% while reducing cost from \$1.17 to \$1.01 per task; on GAIA2, it improves quality from 56.22\% to 62.89\% while reducing cost from \$0.78 to \$0.45 per task. The Pareto-selected harnesses also achieve higher quality at lower cost than quality-only harness optimization. Analysis shows that Pareto-aware optimization retains intermediate candidates that would be discarded by deployment preference but can support later improvements, while Bayesian bootstrap screening filters out candidates whose gains are weakly supported across runs.

cs.AI · cs.CL · cs.LG · cs.MA

版本更新 · v3arXiv:2605.24279

ContextEcho:长程智能体编程会话中人格漂移的基准

ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions

Xianzhong Ding, Yangyang Yu, Changwei Liu, Bill Zhao, Le Chen, Tao Chen

AI 速读 基于摘要

作者提出 ContextEcho 基准与可复用测试框架,用于刻画长程智能体编程会话中编程助手人格的漂移现象。

cs.CL · cs.SE

查看原文摘要

Frontier language models are trained to exhibit stable programming assistant personas, yet those personas may drift over long agentic coding sessions in production. After a long real-world coding session, a model that hedges preferences in a fresh chat ("I don't have preferences") may instead express them directly ("Python -- the feedback loop is instant..."). Existing persona-stability studies focus on scripted chat of tens of turns, leaving long production coding sessions involving thousands of tool-using agent steps, repeated compaction, and hours of interaction largely uncharacterized. We introduce ContextEcho, a benchmark and reusable harness with a 25-probe identity suite, a snapshot-then-probe protocol that evaluates conversation state without perturbing the main session, judge-based and judge-free metrics, and a corpus of 57 long agentic-coding sessions collected with consent and redacted for release across Claude Code and Codex CLI. We use one representative session for a 23-model cross-provider study and all 57 sessions for corpus-scale trajectory analysis across frontier and open-weight models. Results show that persona drift occurs across providers and emerges early in real agentic interaction, while in-session compaction neither reliably resets nor amplifies it. A single-shot anchor restores the original assistant persona, though placebo controls suggest a generic register-reset effect. On same-task continuations, drifted prefixes improve tool-call fidelity -- which we attribute to retained task context rather than drift itself. In tool-free chat, however, the same prefixes break output-format contracts and inflate output length. These findings motivate ContextEcho as an open-source framework for auditing persona drift in long coding sessions without retraining.

版本更新 · v4arXiv:2605.16679

CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?

Haolin Chen, Deon Metelski, Leon Qi, Tao Xia, Joonyul Lee, Steve Brown, Kevin Riley, Frank Wang 等 33 位作者

原文摘要

End-to-end automation of realistic healthcare operations stresses three capabilities underrepresented in current benchmarks: policy density, decisions must be grounded in a large library of medical, insurance, and operational rules; Multi-role composition: a single task requires the agent to play multiple roles with handoffs; and multilateral interaction: intermediate workflow steps are multi-turn dialogs, such as peer-to-peer review and patient outreach. We introduce $χ$-Bench, a benchmark of long-horizon healthcare workflows across three domains: provider prior authorization, payer utilization management, and care management. Each task hands the agent a clinical case in a high-fidelity simulator of 20 healthcare apps exposed via 87 MCP tools, which it must drive to a terminal status through tool calls and writing the role's artifacts, guided by a 1,290+ document managed-care operations handbook skill. Across 30 agent harness/models configurations, the best agent resolves only 28.0% of tasks, no agent clears 20% on strict pass^3, and executing all tasks in a single session slumps the performance to 3.8%. These results raise the hypothesis that similar gaps are likely to surface in other policy-dense, role-composed, irreversible enterprise domains.

cs.CL · cs.AI

新论文 · v1arXiv:2610.36877

GUI-JEPA: Candidate-Conditioned JEPA Latents as a Fast Action Policy for Mobile GUI Agents

Bingchao Wang, Yongcheng Jing, Shunyu Liu, Chengyu Wang, Jun Huang, Jie Yang, Wei Liu, Dacheng Tao

原文摘要

Mobile GUI agents are typically fine-tuned 7B-scale vision-language models (VLMs) that generate each action as text. However, these VLMs decode every action token by token, and even agents that score candidate actions keep a large language model or VLM in the loop, making them slow and memory-hungry. To address these issues, we propose GUI-JEPA, to our knowledge the first decoder-free GUI agent policy built on a Joint-Embedding Predictive Architecture (JEPA), which selects actions in the latent space of a JEPA encoder pretrained on unlabelled GUI screenshots. Its action- and region-conditioned latent predictor turns the screen latent into one latent per candidate action, letting lightweight heads score all candidates against the instruction in a single batched pass, without decoding a token. We further show that the scorer relies on this latent pathway and that the predictor is trained for the decision itself: a randomly initialised predictor does just as well as one pretrained on screen dynamics. On AndroidControl, GUI-JEPA is 2.8-4.0x faster on the same GPU than a Qwen2.5-VL-7B agent we fine-tune with the same inputs, while reaching 86-92% of its in-distribution accuracy with 6% of its parameters. Code is available at https://github.com/bcwang-sjtu/GUI-JEPA

cs.AI · cs.CL · cs.CV

新论文 · v1arXiv:2610.36874

AdaGround: Token-Normalized Adaptive Zooming with Verifier-in-the-Loop Refinement for GUI Grounding

Bingchao Wang, Yongcheng Jing, Chengyu Wang, Jun Huang, Jie Yang, Wei Liu, Dacheng Tao

原文摘要

GUI grounding on high-resolution screens fails mostly on small targets, and zooming in is a common training-free remedy. However, most existing zooming pipelines do not match the zoom to the target's size, accept the zoomed answer unconditionally and never revisit a wrong answer. To address these issues, we propose AdaGround, a training-free framework with three components: (1) Token-Normalized Adaptive Zooming, which sets the magnification, field of view and zoom depth of each sample from the model's own estimate of the target size; (2) Verifier-in-the-Loop Refinement, which uses one marked-click Yes/No verifier to keep or reject each zoom, choose between zoom chains, trigger re-grounding and, with an existence query, decide when to abstain; and (3) Compositional Structure Grounding, which composes spreadsheet cells from their headers. On Qwen3-VL models, AdaGround surpasses state-of-the-art training-free zooming methods and many trained GUI models, improving accuracy over the backbone by 7.4 to 20.6 points on OSWorld-G and ScreenSpot-Pro, and its gains carry over to held-out UI-Vision. Code is available at https://github.com/bcwang-sjtu/AdaGround

cs.CV · cs.AI · cs.CL

版本更新 · v2arXiv:2607.28862

TextCloak: Safeguarding Data from Unauthorized LLM Fine-Tuning via RL-Driven Unlearnable Examples

Chengshuai Zhao, Pingchuan Ma, Dawei Li, Bohan Jiang, Zhiyuan Yu, Zhen Tan, Huan Liu

原文摘要

The development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized online data exploitation and privacy leakage. Unlearnable examples (UEs) emerge as a promising defense by injecting crafted perturbations into data such that models trained on them exhibit degraded utility. However, existing methods for text protection are primarily designed for classification tasks and discriminative language models, which limits their broad applicability (e.g., instruction following or reasoning) in LLMs. In this work, we propose **TextCloak**, an RL-driven data synthesis framework that safeguards textual data against unauthorized LLM fine-tuning. TextCloak employs a *UE generative policy* that transforms clean text into unlearnable examples while preserving semantic fidelity and linguistic naturalness. To optimize the policy, we introduce **GRPO-UE**, which simulates the attacker's LLM fine-tuning and rewards downstream degradation via group-relative policy optimization. This overall formulates a bi-level optimization loop, allowing the generator to discover effective protective patterns. Comprehensive experiments on six publicly available datasets and nine state-of-the-art LLMs demonstrate that TextCloak consistently impairs unauthorized LLM fine-tuning while maintaining text utility for legitimate use. Further analyses confirm its transferability and robustness across model architectures, training configurations, and adaptive attacks, highlighting its broad applicability as a practical defense for online content.

cs.CL · cs.AI · cs.CR · cs.LG

版本更新 · v2arXiv:2601.06086

AzeroS: Extending LLM to Speech with Self-Generated Instruction-Free Tuning

Yiwen Shao, Wei Liu, Jiahong Li, Tianzi Wang, Kun Wei, Meng Yu, Dong Yu

原文摘要

Extending large language models (LLMs) to the speech domain has recently gained significant attention. A typical approach connects a pretrained LLM with an audio encoder through a projection module and trains the resulting model on large-scale, task-specific instruction-tuning datasets. However, curating such instruction-tuning data for specific requirements is time-consuming, and models trained in this manner often generalize poorly to unseen tasks. In this work, we first formulate that the strongest generalization of a speech-LLM is achieved when it is trained with Self-Generated Instruction-Free Tuning (SIFT), in which supervision signals are generated by a frozen LLM using textual representations of speech as input. Our proposed SIFT paradigm eliminates the need for collecting task-specific question-answer pairs and yields the theoretically best generalization to unseen tasks. Building upon this paradigm, we introduce AZeroS (Auden Zero-instruction-tuned Speech-LLM), which is trained on speech-text pairs derived from publicly available corpora, including approximately 25,000 hours of speech with ASR transcripts and 3,000 hours of speech with paralinguistic labels. Built upon Qwen2.5-7B-Instruct, the model updates only two lightweight projection modules (23.8 million parameters each), while keeping both the LLM and audio encoders frozen. Despite the minimal training cost and modest data scale, AZeroS achieves state-of-the-art performance on both semantic and paralinguistic benchmarks, including VoiceBench, AIR-Bench Foundation (Speech), and AIR-Bench Chat (Speech).

cs.CL · cs.SD · eess.AS