Computation and Language
Bridging Semantic Gaps in RAG through Generated Context Knowledge Fusion
Xinkai Du, Chao Lv, Yalin Sun et al. (+3 more)
Retrieval-Augmented Generation has established itself as a fundamental framework in natural
language processing, seamlessly integrating information
retrieval with the generative capabilities of large
language models. However, this process is fundamentally constrained by a critical challenge: semantic space mismatch between queries and retrieved contexts. We propose Knowledge-Aware Semantic Bridging (KASB), a novel framework that improves passage selection quality through semantic space
alignment between queries and retrieved documents through intelligent knowledge fusion. Our approach leverages the complementary strengths of generative and
retrieval-based knowledge through a multistage process that enhances both relevance and accuracy. We evaluate KASB on three popular open-domain Question Answering
datasets to demonstrate the effectiveness of our approach.
Read the full abstract ▾
All authors Xinkai Du, Chao Lv, Yalin Sun, Quanjie Han, Lei Yao, Maosong Sun
Computation and Language · Cross-listed from cs.LG
Trajectory Soup: Pushing the Compute-Scaling Frontier of LLM Mid-training via Diverse Trajectories
Zhehao Huang, Changxin Tian, Qingyuan Yang et al. (+5 more)
Mid-training equips pretrained large
language models with specialized and
reasoning capabilities, but the returns of this stage are bounded since additional serial compute yields little further downstream improvement and can even degrade some capabilities, which places a practical ceiling on how much compute mid-training absorbs. We revisit how this compute should be allocated to a single run or multiple similar optimizations. We find that branches forked from a shared checkpoint under various controlled recipe reaches measurably different regions of parameter space, and establish a form of compatible diversity that extending one run cannot supply. Therefore, we introduce Trajectory Soup, which distributes a mid-training budget over several independent branches, and consolidates strongest checkpoints selected on validation through intra- and inter-trajectory averaging into a single model. A local bias and variance analysis separates the two averaging levels, showing that inter-trajectory averaging removes residual error beyond the reach of averaging within a trajectory, while checkpoint selection carries a bias that bounds how many checkpoints are worth merging. Across model scales, learning-rate schedules,
token budgets, and trajectory counts, Trajectory Soup improves aggregate downstream performance over the strongest single-trajectory average under matched budgets and keeps improving as budgets expand, with the advantage preserved after an identical post-training pipeline. These results position trajectory allocation and merging as a practical way to extend the compute-scaling frontier of mid-training beyond serial saturation.
Read the full abstract ▾
All authors Zhehao Huang, Changxin Tian, Qingyuan Yang, Kunlong Chen, Ziqi Liu, Zhiqiang Zhang, Xiaolin Huang, Jun Zhou
Computation and Language · Cross-listed from cs.LG
Multimodal Detection of Higher-Order Behavioral Constructs: Self-Compassion in Structured Reflective Interaction
Siddhant Jain and Dimitra Tsovaltzi
Many of the qualities that matter most in how people learn and grow, how someone regulates their emotions, reflects on a setback, or stays aware of others during a difficult conversation, are not directly observable. They have to be inferred from how someone speaks, moves, and sounds over time, and they resist the kind of clean labeling that most machine learning pipelines are built around. We study this challenge through a case that is well grounded in psychological theory but rarely modeled computationally: self-compassion, the tendency to respond to one's own setbacks with patience rather than harsh self-criticism. We examine how it appears during structured reflective interviews in a technology-mediated training setting, where people naturally talk through socio-emotionally demanding situations. Since no existing
dataset captures this kind of construct in this kind of setting, we collected and annotated 51 reflective dialog sessions using an independent, temporally overlapping annotation scheme grounded in established theory. We consolidate the underlying six-component psychological model into a three-class supervision space, balancing self-kindness and mindfulness against self-critical or overwhelmed states, and build a reproducible window-based pipeline that aligns
video, audio, and text on a shared timeline. Unimodal models trained on each modality separately are compared against a simple probability-level fusion strategy, which yields modest but consistent gains over the best single modality. We close by discussing where each modality succeeds or struggles, what this suggests about how this kind of construct is actually expressed in reflective
speech, and what would be needed to model it, and constructs like it, more effectively.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.SE
LoLBench: Evaluating Coding Agents with Long-Horizon Proposals on Large Software Systems
Yun Peng, Zihan Wu, Zeyang Zhuang et al. (+6 more)
Modern coding
agents can deliver increasingly large repository-level changes, and recent
benchmarks reflect this by emphasizing long-horizon tasks with large reference implementations. Many
benchmarks evaluate coding
agents' implementation capability to produce correct code edits from detailed specifications. However, practical modular development tasks also require the perception capability of grounding user intent and high-level design to derive a specification. We introduce LoLBench to evaluate both capabilities through the entire proposal-to-implementation process on large software systems. It is a multilingual
benchmark of 100 tasks across 29 software systems in five domains. Each task provides a human-written enhancement proposal with user intent and high-level design. On average, proposals contain about 5,000 words, software systems contain 2.4 million source lines of code (LoC), and implementation pull requests (PRs) change approximately 5,500 LoC. Across 28
agents we evaluated, the best
agent resolves only 14% of tasks and achieves a 52.7% Fail-to-Pass (F2P) pass rate. Failure analysis identifies incomplete code localization as a major bottleneck, while providing reference-derived file trees alongside API specifications improves resolved rates by 16–22 percentage points (2.4–17$\times$), reaching at most 34%. These results show that both perception and implementation remain central challenges for coding
agents in practical modular development on large software systems. LoLBench is available at https://huggingface.co/
datasets/lolbench26/LoLBench.
Read the full abstract ▾
All authors Yun Peng, Zihan Wu, Zeyang Zhuang, Xin Zhou, Rui Shu, Xu Han, Chun Yong Chong, Yuan Wang, Jiakun Liu
Computation and Language
LLM unbranding: Erasing Commercial Identity while Preserving Generic Utility
Kajetan Ożóg, Alicja Wojciechowska, Dawid Malarz et al. (+3 more)
Establishing unbranding as a critical practice to prevent visual logos from acquiring negative connotations is standard in image generation. Large
Language Models (LLMs) now face a parallel and emerging challenge. These models frequently generate brand descriptions within diverse contexts. This frequency introduces significant risks, such as trademark dilution, false attribution, and brand defamation. In response, we formally define the novel task of LLM Unbranding. We specifically address the complex challenge of managing trade dress within textual outputs. This involves neutralizing characteristic
language, slogans, and stylistic markers that define brand identity. Crucially, these elements are less evident than explicit visual logos. To
benchmark this task, we introduce a comprehensive evaluation
dataset incorporating prominent brands from multiple commercial domains. We rigorously evaluate existing state-of-the-art machine unlearning models using this
benchmark. This evaluation identifies their limitations in selective textual unbranding. Finally, we propose MUTE, a novel inference-time method that effectively neutralizes textual trade dress while preserving the LLM's general capabilities and utility. By leveraging an iterative refinement loop, MUTE systematically optimizes system instructions to safely eliminate brand leakage without requiring fragile parameter updates. Code and
dataset: The evaluation
dataset and code for LLM Unbranding are available at https://github.com/KajetanOzog/LLM_unbranding. The implementation of MUTE is available at https://github.com/KajetanOzog/MUTE.
Read the full abstract ▾
All authors Kajetan Ożóg, Alicja Wojciechowska, Dawid Malarz, Paweł Batorski, Artur Kasymov, Przemysław Spurek
Computation and Language
Cross-Linguistic Effects in Bilingual Phoneme BabyLMs
Nikitas Theodoropoulos, Maria Lymperaiou and Giorgos Filandrianos
Cross-linguistic effects are a central topic in bilingual first-
language acquisition. Artificial learners can help investigate L1-L2 interactions by enabling controlled comparisons across
language combinations and learning conditions. Recent work explores this direction by training bilingual
language models under developmentally plausible constraints. However, human and model learners still diverge in fundamental ways, with one major difference being input modality: children learn primarily from spoken input, whereas
language models are typically trained on orthographic text. To reduce this gap, researchers have trained models on phonemic representations of
speech. In this work, we combine these research directions to train bilingual BabyLMs with phonemic input. We keep English fixed as the L2 and vary the L1 across German, Swedish, Persian, and Basque, selected to represent contrasting combinations of syntactic and phoneme-inventory distance from English. Our results show stronger L1-related variation in grammatical learning trajectories under phonemic than orthographic input, while early lexical differences align with phoneme-inventory similarity.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.LG
Unlocking the Critic: Reward-Free Policy Optimization for LLM Post-Training
Hongyang Li, Xiao Li, Caesar Wu et al. (+3 more)
Recent approaches to reinforcement learning (RL) post-training for large
language models increasingly remove the critic to reduce training instability and
memory overhead. Even where a critic is trained, it is discarded once training ends, although it has learned to predict outcomes. We revisit this trend and show that a pretrained critic's ability to predict future outcomes can make it a valuable asset for efficient long-horizon
reasoning. First, we find that instability in critic-based RL for long chain-of-thought
reasoning is largely an optimization artifact: keeping policy updates small and low in variance restores stable convergence. Second, a well-pretrained critic estimates the posterior probability of eventual success from later trajectory states and unfinished prefixes. Its predictions provide outcome-derived, dense, per-prefix learning signals that, during policy optimization, require neither completed rollouts, step-level annotations, nor external reward labels. Building on this insight, we introduce Reward-Free Policy Optimization (RFPO), which repurposes a single calibrated, frozen critic as a rollout-level reward, a value baseline for generalized advantage estimation, and a success forecaster for unfinished prefixes. We further show that binarizing the debiased score stops the policy from exploiting the critic's length bias. Binarized, RFPO matches supervised PPO without a single label in the training loop, while cutting compute and
memory overhead. This makes RFPO well suited to long-horizon
reasoning tasks, where outcomes arrive late and generation dominates cost: because rollouts can be rewarded before they finish, training no longer has to pay for waiting on every trajectory to complete. Our findings challenge the prevailing critic-free paradigm and establish critic-based, reward-free optimization as a scalable and computationally efficient path for LLM post-training.
Read the full abstract ▾
All authors Hongyang Li, Xiao Li, Caesar Wu, Said Mammar, Grégoire Danoy, Pascal Bouvry
Computation and Language · Cross-listed from cs.LG
VACE: Validation-Gated Alternating Co-Evolution of Agent Models and Harnesses
Jiexing Qi, Yu He, Jun Liu et al. (+9 more)
Language model
agents can be improved by updating their model weights or refining the harness that guides task execution. These components are coupled: weight updates change how the model uses the harness, while harness updates change the trajectories used for training. We propose VACE, Validation-Gated Alternating CoEvolution, which alternates agentic reinforcement learning with trajectory-driven harness refinement. After each RL stage, VACE reuses the collected trajectories to propose a harness revision and evaluates the incumbent and candidate with the updated model held fixed. The candidate guides subsequent training only if it improves validation performance. With Qwen3.5-9B, VACE achieves 45.26% test accuracy on OfficeQA and a mean partial-credit score of 75.19% on AutomationBench, exceeding weight-only RL by 6.43 and 9.09 percentage points and ungated alternation by 4.59 and 6.95 points, respectively. Across 44 harness proposals, 17 reduce validation performance at the updated checkpoint and are rejected before subsequent RL training, highlighting the importance of validation gating.
Read the full abstract ▾
All authors Jiexing Qi, Yu He, Jun Liu, Qichen Huang, Shaohua Hu, Zhan Dang, Guohua Chen, Rui Yang, Wen Jiang, Yang Liu, Tao Lyu, Fangming Li
Computation and Language
What Does Post-Training Change in Multilingual Reasoning?
Hongyang Li, Xiao Li, Caesar Wu et al. (+2 more)
Open-source
reasoning models provide unequal access to
reasoning capability across
languages. When a model can solve a problem but cannot deliver a complete solution in the user's
language,
language becomes an access barrier rather than merely a source of performance variation. We audit Qwen3 checkpoints on competition-mathematics tasks in eleven
languages. Across the ten non-English
languages, only 15.4-17.9% of problems receive a correct, terminating solution with visible
reasoning in the requested
language in any of 16 samples, compared with 92.9% in English. To identify the source of this disparity, we evaluate thirteen endpoints from one model family, spanning released checkpoints, multilingual supervised fine-tuning (SFT) at two scales, controlled SFT ablations, and three reinforcement-learning (RL) reward formulations. We jointly track correctness,
language adherence, termination, and delivery efficiency. The dominant bottleneck shifts across post-training stages. Released models often reason in English. Multilingual SFT restores target-
language reasoning, but accuracy declines across multilingual, English-only, and single-
language SFT runs, showing that this cost is not specific to multilingual mixing; non-English
reasoning traces additionally become prone to non-terminating loops. RL restores termination in both arms at no cost in accuracy, but only the arm whose reward includes a
language term delivers: rewarding correctness alone returns the model to English. Together, these stages establish a constructive post-training path from English-pivoted capability to multilingual
reasoning that is reliably delivered.
Read the full abstract ▾
All authors Hongyang Li, Xiao Li, Caesar Wu, Grégoire Danoy, Pascal Bouvry
Computation and Language
Traverse: Learning When to Remember, Reset, and Redirect for Long-Horizon Web Search
Jingyuan Ma, Lynx Aster, He Zhang et al. (+5 more)
Long-horizon information-seeking
agents often accumulate noisy or misleading context, causing early mistakes to persist and making recovery increasingly difficult. We introduce an autonomous search harness in which the
agent manages its own search process through three states: Rubric, Answer, and Verify. The
agent first defines criteria for a valid answer, searches under these criteria, and then independently verifies the result before deciding whether to terminate or continue searching. It is further equipped with a Seal
Memory tool that enables active context management. Training this behavior with reinforcement learning, however, can induce Seal Collapse, resulting in unstable training and preventing the
agent from reliably learning when and how to use its
memory tools. We solve this with a simple strategy that trains only the final segment after context management. Our 35B model achieves 72.83 on BrowseComp, outperforming comparable open-source systems, and consistently improves over the base model across BrowseComp-ZH, xbench, DeepSearchQA, WideSearch, financial investigation, and product search. Ablations show that autonomous compression outperforms automatic compaction and validate our RL design.
Read the full abstract ▾
All authors Jingyuan Ma, Lynx Aster, He Zhang, Siyao Song, Weijie Yuan, Zhe Zhang, Kai Jia, Zhifang Sui
Computation and Language
Learning from Think-Mode Advantage via On-Policy Distillation
Wanqi Ren, Jianxiang Wang, Danxuan Liu et al. (+2 more)
Explicit intermediate
reasoning gives large
language models (LLMs) a stronger problem-solving mode. We study learning from this think-mode advantage via on-policy
distillation (OPD). OPD preserves student-generated trajectories and provides dense
token-level teacher targets at student-visited prefixes. Privileged
reasoning is used during
distillation rather than student inference. Uniform ThinkOPD, a natural think-enabled OPD baseline, conditions a fixed teacher on one shared think trace and uniformly distills every sibling student response. Although its prefixes are on-policy, the trace need not follow a route compatible with every complete response: the same privileged trace can induce different teacher-student discrepancies even when responses reach the same outcome. We summarize this interaction with trace-response divergence (TRD) and introduce ThinkOPD, which routes supervision at the response level by combining group-relative reward gain with a TRD-based compatibility proxy. Final response weights are normalized within each rollout group. Across mathematical
reasoning and code generation, ThinkOPD outperforms Uniform ThinkOPD in both same-model settings and both cross-model teacher-student pairs, and it exceeds representative rationale and self-
distillation baselines in a controlled comparison. Controlled interventions show that outcome benefit and the TRD-based proxy provide complementary routing signals in this setting. Think-enabled OPD provides a controlled setting for studying how teacher advantage becomes transferable along student responses.
Read the full abstract ▾
All authors Wanqi Ren, Jianxiang Wang, Danxuan Liu, Linyi Ding, Huaixiao Tou
Computation and Language
Selecting The Most Informative Tokens in Natural Language Autoencoders
Federico Torrielli, Gianluca Barmina, Andrea Blasi Núñez et al. (+4 more)
Natural
language autoencoders translate a
language model's internal activations into readable explanations. Explaining every
token position is costly. Which positions should an auditor inspect to understand a potential threat? We study this question across $4.7$ million explanations on
prompt injection and concealment. We compare signals from model computation with a ranker trained only on chat structure. Chat structure usually selects more relevant explanations than the computational signals, without requiring a model forward pass for position selection. On three of four
datasets, explaining just $5\%$ of positions retains nearly all of the success rate from explaining every position, where success means obtaining an explanation about the threat. The benefit varies with the audit task. We also show that pretrained verbalizers recover words that models have learned to conceal through fine-tuning, without additional verbalizer training. These results identify where auditors can concentrate explanation generation and show that useful explanations can extend beyond the model a verbalizer was trained to describe.
Read the full abstract ▾
All authors Federico Torrielli, Gianluca Barmina, Andrea Blasi Núñez, Amon Rapp, Luigi Di Caro, Peter Schneider-Kamp, Lukas Galke Poech
Computation and Language
LatCom: Cross-Agent Latent Compression for Efficient Multi-Agent Collaboration
Shinan Zhang, Tao Zhang, Qihui Zhu et al. (+7 more)
LLM-based multi-
agent systems (MAS) increasingly use latent collaboration to avoid the information loss and repeated encoding-decoding overhead of natural-
language communication. However, directly forwarding all sender latents makes the receiver-side context scale with both the number of
agents and the
reasoning length, increasing computation,
memory usage, and collaboration latency. A natural solution is latent compression. But we find that cross-
agent redundancy remains unresolved in existing latent compression approaches, which typically compress each sender independently and then concatenate the results. We propose LatCom, a cross-
agent latent compression framework for efficient multi-
agent latent collaboration. LatCom maps multiple sender latents into a fixed number of receiver-readable and task-relevant slots. Rather than reconstructing all sender hidden states, it optimizes the compressed latents for receiver-side task utility. LatCom trains the compressor in two stages: single-sender readability learning first establishes a latent interface interpretable by the frozen receiver, and multi-sender fusion learning then trains the compressor to fuse complementary evidence and remove redundancy across
agents. Experiments on multiple
benchmarks with Qwen3-4B show that LatCom achieves an average 2.46x inference speed-up over LatentMAS and reduces output
token usage by 70.3% while maintaining comparable average accuracy.
Read the full abstract ▾
All authors Shinan Zhang, Tao Zhang, Qihui Zhu, Mengjie Zhang, Dong Jin, Yunpeng Hou, Shuangwu Chen, Xiaobin Tan, Quan Zheng, Jian Yang
Computation and Language
CypherTurn: A Multi-Turn Benchmark for Conversational Text-to-Cypher Evaluation and the Autonomy Divergence
Yuzhe Zhang, Weijie Zhu, Haolin Yang et al. (+5 more)
Graph databases are increasingly queried through natural
language, yet every existing
benchmark evaluates isolated single-turn queries rather than the multi-turn sessions through which analysts actually work. We introduce CypherTurn, the first
benchmark for conversational Text-to-Cypher evaluation, comprising 721 sessions and 5,927 turns across 7 knowledge graphs and 13 conversational phenomena. We evaluate 15 models under a guided oracle protocol and a fully autonomous agentic protocol, yielding four findings. First, the best model reaches only 64.7% execution accuracy, and session-level correctness remains below 5%. Second, despite strong overall rank correlation, frontier models exhibit a consequential reordering of the top of the leaderboard under autonomous operation, a phenomenon we term the Autonomy Divergence, which reveals error-management as a partially independent capability from raw generation skill. Third, scaling action budgets from x3 to x10 fails to close the autonomy gap, as the strongest frontier models self-limit to approximately two actions per turn regardless of available budget. Fourth, single-turn Cypher fine-tuning degrades multi-turn instruction following, while architecture-appropriate specialization outperforms several frontier models. These results establish CypherTurn as an open challenge for conversational graph database
reasoning. Code and data are available at https://github.com/BarryQ/CypherTurn.
Read the full abstract ▾
All authors Yuzhe Zhang, Weijie Zhu, Haolin Yang, Ziyun Zhang, Xianwei Xue, Mengke Chen, Qiutong Pan, Huaqian Cai
Computation and Language
SRJudge: Empowering Large Language Models with Selective Reasoning for Fine-Grained Knowledge Concept Tagging
Zhiwei Yang, Jiahua Yang, Huiru Lin et al. (+2 more)
Knowledge concept tagging aims to assign specific concept or topic labels to educational content, which is essential for both educators and learners in traditional and online teaching practices. Recent work has explored large
language models (LLMs) for this task, achieving promising performance. However, LLMs still struggle to select the correct concept from a large-scale candidate set due to the high dimensionality of the decision space. In this paper, we propose a novel three-stage Select-Reason-Judge (SRJudge) framework, which empowers LLMs with selective
reasoning capability for fine-grained knowledge concept tagging. Specifically, the Selector in Stage 1 first narrows the candidate concepts to a top-K shortlist by fine-tuning a small
language model (SLM), e.g., BERT, since the top-$K$ predictions hit the correct concept in most cases, thereby reducing the decision space of correct candidates. Next, the Stage 2 Reasoner employs a lightweight LLM for refined
reasoning over the shortlisted candidates. It further integrates an improved reinforcement learning strategy with a dynamic task-specific reward function and a
pruning mechanism to better align with human
reasoning preferences. Finally, a larger LLM acts as a judger that evaluates the overall rationality of the
reasoning process and its explanations to determine the final output. In addition, we construct two high-quality
datasets for further validation, i.e., the biology
dataset S_Bio and the physics
dataset S_Phy. Experimental results demonstrate that our method consistently outperforms state-of-the-art baselines across
benchmark datasets, verifying its effectiveness and superiority. Resources are available at: https://github.com/Nicozwy/SRJudge.
Read the full abstract ▾
All authors Zhiwei Yang, Jiahua Yang, Huiru Lin, Xing Chen, Quanlong Guan
Computation and Language
AMU:Admission and Memory Update for Personalized Conversations—Structured Memory with SLM Guided Control
Tao Hwang and Yishi Diao
Large
language models (LLMs) have become the foundation of personalized assistants, but maintaining persistent user
memory across long-term interactions remains challenging. Existing
memory systems often focus on storage,
retrieval, or consolidation, while
memory writing remains less controlled: transient requests, duplicate statements, and outdated user states may enter
memory and later be retrieved for personalization. In this paper, we present AMU: Admission and
Memory Update for Personalized Conversations, an SLM-guided (Small
language model guided) structured framework for writing-time
memory control. AMU uses structured
memory filtering to decide what should enter
memory and SLM-guided storage management to determine whether an admitted record should be stored separately, discarded as a duplicate, or fused as an update. We evaluate AMU in a controlled
memory writing and
retrieval setting. Experimental results show that AMU maintains cleaner and more retrievable personalized
memories.
Read the full abstract ▾
Computation and Language
Repetition, Not Length: Isolating the Counting Failure in Neural Text-to-Speech
Kirill Borodin, Vasilii Kudryavtsev, Maxim Maslov et al. (+1 more)
Text-to-
speech models loop, truncate and lose count on text that repeats a phrase many times. We show that repetition itself is what breaks them, not the length that comes with it. Every repeated sentence in our test set is paired with a control of matched sentence and word count in which no word ever repeats back-to-back. Six models from three architectures render the controls almost perfectly and fail the repeated twins: 94.3% against 18.2% exactly right at k >= 6. The gap survives greedy decoding, repetition-penalty sweeps, four independent
speech recognisers and 420 analysis specifications without once reversing sign; a held-out fourth architecture lands within a point of its predicted gap, and one of two non-autoregressive baselines shows the same failure. Varying the period of the text shows the failure grows smoothly with periodicity, half of it surviving when no word is adjacent to itself.
Read the full abstract ▾
All authors Kirill Borodin, Vasilii Kudryavtsev, Maxim Maslov, Grach Mkrtchian
Computation and Language
Chinese-Jev: Bringing System One Model to Chinese-Language Tasks
Zexiao Wang, Zihao Zhang, Xudong Wang et al. (+5 more)
System One models such as Jev offer an efficient alternative to generative
language models for tasks that require decisions rather than open-ended responses. However, existing Jev models exhibit limited Chinese-
language decision accuracy, restricting their utility in both general and specialized settings. In this paper, we introduce Chinese-Jev, a System One model that addresses this gap through a unified data processing and training pipeline. Our data processing protocol converts heterogeneous Chinese-
language annotations into probability targets over candidate options, enabling a shared training formulation across domains and question formats. To enable efficient inference, Chinese-Jev adopts a lightweight encoder-only backbone for text encoding and learns to score candidate answers through decision-oriented training. To address the misalignment between the pre-training distribution and downstream Chinese-
language scenarios, we first train the model on a general-purpose corpus of 10 million examples, then fine-tune it separately for the medical, legal, and financial domains. To evaluate decision accuracy and calibration in both general and domain-specific Chinese-
language settings, we introduce Chinese-Jev Bench (CJ-Bench). After first-stage pre-training, Chinese-Jev exceeds the accuracy of the closed-source Jev model by 1.24% on general-domain tasks while achieving a 20.3x speedup. Subsequent domain-specific fine-tuning yields a 4.0% accuracy improvement over Jev in medicine and achieves 92% of Jev's average accuracy across specialized domains, with a 17x speedup and an average latency of only 15 ms per example. We further demonstrate on-device deployment of an INT8-quantized model on mobile devices, achieving an inference latency of approximately 1.0 second per decision. The project is available at https://gulucaptain.github.io/Chinese-Jev/.
Read the full abstract ▾
All authors Zexiao Wang, Zihao Zhang, Xudong Wang, Pan Wang, Ziyi Ye, Haoyu Zhao, Zuxuan Wu, Shuicheng Yan
Computation and Language · Cross-listed from cs.LG
VStress: Correlation-Aware Auditing and Adaptive Budget Allocation for Repeated Verifiers
Miaobo Hu, Shuhao Hu, Xiaobo Guo et al. (+5 more)
Repeated verifier calls are useful only when they contribute conditional information. We introduce VStress, an auditable replay contract, and VStress-CA, a correlation-aware allocation policy that estimates the conditional marginal information of an unqueried verifier on a sealed calibration split, discounts uncertainty, normalizes by call cost, and stops or abstains when the next call is not informative. The controller freezes its decision and cost ledger before joining the clean oracle; a dependence-shift alarm disables channel preference and falls back to exact-stop. The controlled audit gives the mechanism boundary: at 35% symmetric corruption, majority-5 improves balanced accuracy from 0.6578 to 0.7739, whereas at 65% it loses 0.1226 points. In the matched fixed-budget comparison, breadth, redundancy, and adaptive allocation obtain balanced accuracies 0.6048, 0.6375, and 0.6538, with 3.4216 calls per item and an RLVR score of 0.6417 for VStress-CA. Dependence diagnostics also increase from same-model repeats to cross-family channels, with conditional marginal gains of 0.0126, 0.0462, and 0.0913. These measurements turn correlation from a post-hoc warning into an auditable allocation decision.
Read the full abstract ▾
All authors Miaobo Hu, Shuhao Hu, Xiaobo Guo, Xin Wang, Bokun Wang, Peng Zhang, Daren Zha, Jun Xiao
Computation and Language · Cross-listed from cs.LG
Cool the Sampler, Not the Learner: Sampling Temperature Moves the Staleness Cliff of Importance-Corrected GRPO
Taiheng Pan
Production RL for
language models lets the sampler fall behind the learner and repairs the resulting mismatch with a truncated importance weight. We ask how long the sampler can go without a refresh under that correction, and find a cliff: on Qwen2.5-Math-1.5B and GSM8K, importance-corrected GRPO refreshed every 192 updates learns well for 180 steps and then degrades severely in all three data seeds before the refresh arrives. Published remedies for staleness act on the update; we act on the sampler instead. Decoupled cooling draws samples at temperature 0.8 while the learner, the reference model and the importance weights stay at temperature 1, with the behaviour probability recorded from the tempered distribution, so the learner's objective is unchanged. All corresponding cooled runs are stable, and the longer interval keeps what the short one delivered: at the same update budget, a cooled sampler refreshed every 192 steps matches an uncooled sampler refreshed every 96 at the end of training (0.857 for both) and averaged over it (0.79), whereas lowering the learning rate to a safe value ends 3-7 points lower. On Qwen2.5-Math-7B the degradation points at interval 192 predict that an interval of 144 is fatal without cooling and survivable with it; on two data seeds the uncooled runs degrade before their first refresh and the cooled runs pass it and end at 92-93% against 68-81%, with one cooled run degrading transiently late in the second cycle. The benefit has a window: at three times the safe interval and in a high-mismatch MATH setting cooling delays degradation without preventing it, stronger cooling is not better, and cooling without the correction collapses. Sampling temperature is a control on staleness tolerance, and temperature and refresh interval should be chosen together.
Read the full abstract ▾
Computation and Language
ER-JEPA: Experience Replay Improves Joint-Embedding Predictive Learning in Language Models
Jingnan Pu, Zi-En Fan and Feng Lian
Large
language models (LLMs) excel at
token-level generation but may learn undesirable abstract semantics and lack comprehensive perception. LLM-JEPA mitigates this by aligning different views of the same underlying knowledge via a joint-
embedding predictive architecture (JEPA). However, strong
alignment does not necessarily lead to accurate, stable predictions. To address this, we propose ER-JEPA, which adds an episodic replay path to LLM-JEPA. ER-JEPA stores training pairs in a
memory. At each step, it stores and retrieves relevant data to provide additional supervision. This enables learning from both the current batch and stored training pairs, providing additional supervision for
token prediction and representation
alignment. Experiments across multiple
datasets (NL-RX, GSM8K, Spider, and NQ-Open) demonstrate that ER-JEPA consistently outperforms LLM-JEPA.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.AI
CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory
Jingguang Li, Yebo Wu, Zuyi Guo et al. (+7 more)
Long-context
reasoning is essential for complex and long-horizon tasks, yet the performance of large
language models (LLMs) degrades as context length increases. Recent approaches address this by processing input chunk by chunk while maintaining a bounded textual
memory in model context. However, premature information compression can discard critical details essential for subsequent
reasoning. In this paper, we introduce Commit-on-Evidence
Memory (CoEM), which learns when to convert source evidence into compact
memory facts. Specifically, under a fixed context-
memory budget, CoEM preserves potentially useful source excerpts verbatim in a pending set, allowing subsequent context to clarify their relevance before irreversible compression. As new context arrives, a learned policy revisits each pending excerpt and decides whether to promote it to the committed
memory, retain it for further consideration, or discard it. A frozen verifier ensures proposed facts are accepted only if supported by retained excerpts and current context. To further guide effective
memory management, we train this policy using reinforcement learning by combining fine-grained, step-level evidence rewards with final answer rewards. Extensive experiments demonstrate that CoEM consistently improves long-context
reasoning. When evaluated on 6,400 documents long-context input, CoEM outperforms the strongest
memory baseline by 10.4-11.4 F1 points on Qwen3.5-9B. Code repository: https://github.com/benmagnifico/CoEM.
Read the full abstract ▾
All authors Jingguang Li, Yebo Wu, Zuyi Guo, Kailang Ma, Xianjie Dai, Han Zheng, Benwang Chen, Li Li, Can Rong, Heye Huang
Computation and Language
Dating the Model: Hidden Dates in System Prompts Affect LLM Evaluation
Mario Sanz-Guerrero, Minh Duc Bui, Manuel Mager et al. (+1 more)
Reproducibility is essential for scientific research, yet prior work shows that LLM outputs vary with hardware and batching. We identify an overlooked factor: the hidden injection of the current date into system
prompts, which users cannot control and which changes every day. Across 9 recent LLMs and 6
datasets spanning multiple-choice QA (MCQA), math
reasoning, code generation, and machine translation, performance varies solely with the current date, with deltas of up to 6% on MCQA, 14% on math
reasoning, 7% on code generation, and 2.84 BLEU on machine translation. Model rankings also shift, affecting leaderboards. This date effect exceeds other sources of non-determinism, such as batch size and numerical precision. Standard prompting techniques – chain-of-thought and few-shot prompting – do not reduce the sensitivity; chain-of-thought even amplifies it. Our findings underscore the need for careful evaluation protocols to ensure reproducibility and fair comparisons in LLM research.
Read the full abstract ▾
All authors Mario Sanz-Guerrero, Minh Duc Bui, Manuel Mager, Katharina von der Wense
Computation and Language
Benchmarking Automatic Speech Recognition Tools for Iberian Languages
Fernando López, Pablo Gómez, David Solans et al. (+2 more)
Comprehensive evaluations of automatic
speech recognition (ASR) for Iberian
languages remain limited, and low-resource
languages, biases, and efficiency trade-offs are underexplored. We
benchmark eleven systems, ten open-weight models and one commercial API, across five Iberian
languages (Basque, Catalan, Galician, Portuguese, Spanish), with German and Turkish as controls. Evaluation uses an 85-hour
dataset covering read
speech, broadcast media, and audiobooks, assessing accuracy and efficiency via word error rate (WER) and real-time factors (RTF/RTFx). Results show no single model dominates: accuracy, efficiency, and
language coverage present clear trade-offs. Low-resource
languages, especially Basque, degrade significantly, highlighting the role of training coverage. We observe consistent sex disparities across most systems, highlighting
fairness challenges in multilingual ASR. Overall, the
benchmark provides practical guidance for real-world model selection.
Read the full abstract ▾
All authors Fernando López, Pablo Gómez, David Solans, Paulo Villegas, Jordi Luque
Computation and Language
Can Language Models Learn to Forecast Stock Prices
Jiacheng Guo, Suozhi Huang, Shuzhen Li et al. (+11 more)
Post-training has been shown to significantly improve
language models' performance on tasks with verifiable outcomes, including mathematical
reasoning, software engineering, and computer use. However, whether the same approach can improve forecasting in financial
markets is much less clear. Compared with tasks with verifiable outcomes, not only are realized returns noisy, but even what constitutes a relevant information set for making effective predictions is not obvious a priori: the model must decide which observations to gather and then commit to a numerical judgment before the outcome is known. We study this question in a chronological stock-price sandbox, where a
language model gathers price, volume, relative-performance, and
market-context evidence and predicts a future return. We post-train Qwen3-4B with supervised fine-tuning (SFT) on tool-use demonstrations, then proximal policy optimization (PPO) with a terminal reward given by the forecast score against the realized return. The resulting AURA-4B more than doubles the starting direction–magnitude score, from 20.94 to 43.31, and is comparable to frontier
language models on this
benchmark. Conditional magnitude agreement rises from 33.3 to 66.2, while directional accuracy changes from 62.9 to 65.4. SFT expands tool use, and PPO further increases the share of ranking and
market-context queries. These results show that post-training can substantially improve financial forecasting performance, together with changes in how the model investigates the
market, on this outcome-selected
benchmark.
Read the full abstract ▾
All authors Jiacheng Guo, Suozhi Huang, Shuzhen Li, Yunlong Gao, Zerui Cheng, Jason Ge, Shushu Liang, Zihao Li, Hao Lu, Ming Yin, Shilong Liu, Jiashuo Liu, Xu Kuang, Mengdi Wang
Computation and Language
BaLEEN: Biasing with Latent Encoded Entities for Context-Aware ASR
Chihiro Taguchi, Yotaro Kubo and Rujikorn Charakorn
Transcribing domain-specific entities and rare proper nouns remains a major challenge in automatic
speech recognition (ASR). In this paper, we propose BaLEEN (Biasing with Latent Encoded Entities), a lightweight, hypernetwork-based framework for dynamic contextual adaptation without fine-tuning the underlying ASR model. BaLEEN encodes variable-length contextual keywords using a pretrained
language model, compresses them into a fixed sequence of latent vectors via a Perceiver bottleneck, and injects context-dependent bias vectors directly into the intermediate encoder representations of the ASR model. Because both the
language model and the backbone ASR model remain entirely frozen during training, BaLEEN operates as a plug-and-play adapter that incurs zero computational overhead at inference time when context biases are precomputed. We evaluate our method on a CTC-based ASR model using a Wikipedia-derived corpus with annotated named entities and synthetic
speech. Experimental results demonstrate that BaLEEN reduces keyword miss rate by 8.7% on the test set relative to the unbiased baseline while simultaneously improving overall word error rate by 21% and character error rate by 28%.
Read the full abstract ▾
Computation and Language
MultiTalk: Scaling Full-Duplex Speech Models to Long, Multi-Party, Bilingual Conversation
Ke Wang, Houxing Ren, Zimu Lu et al. (+4 more)
End-to-end full-duplex
speech models have brought open-source machine conversation closer to human-like interaction, yet existing systems remain limited in two intertwined dimensions: long-context robustness and multi-party interaction. Real-world scenarios such as meetings, group lessons, and social-
robot reception require a single model to track, contextualize, and respond to multiple speakers over extended durations. Progress is constrained by both data and evaluation: open multi-party
speech corpora remain small and are not designed for codec-frame-level full-duplex modeling, while existing long-audio
benchmarks focus on passive listening and
speech-to-
speech benchmarks are mostly short and dyadic. We extend the Moshi paradigm jointly along the long-horizon and multi-party axes in English and Chinese. First, we release 57.6k hours of synthetic training data ($\href{https://huggingface.co/datasets/MultiTalk/MultiTalkPT}{MultiTalkPT}$ and $\href{https://huggingface.co/datasets/MultiTalk/MultiTalkFT}{MultiTalkFT}$) for long-form, multi-party, English-Chinese full-duplex dialogue, with controllable length, participant count, turn-taking, overlap, backchannels, interruptions, addressee shifts, and long-range coreference. Second, we introduce $\href{https://huggingface.co/datasets/MultiTalk/MultiTalkBench}{MultiTalkBench}$, built from real human recordings, for evaluating long-form, multi-party, bilingual full-duplex dialogue. Conversations average 32.6 minutes and include probes for long-range entity tracking, topic coherence, and addressee selection. Third, we train a bilingual Moshi-style model that sustains coherent multi-party English-Chinese conversations over extended durations and substantially outperforms open-source baselines including Moshi, MiniCPM-o-4.5, and Qwen3-Omni-30B-A3B-Instruct on MultiTalkBench.
Read the full abstract ▾
All authors Ke Wang, Houxing Ren, Zimu Lu, Yunqiao Yang, Zhuofan Zong, Mingjie Zhan, Hongsheng Li
Computation and Language
RAEGNet: Relation-Aware Evidence Graph Network for Harm-Aware Multimodal Fake News Detection
Wenbin Shen, Guoxuan Qin, Guangxu Yao et al. (+4 more)
Existing
multimodal fake news detection methods often introduce external information to assist detection. However, most of them rely on entity-level
retrieval and are therefore prone to introducing event-irrelevant noise. Meanwhile, existing methods mainly focus on improving overall performance and do not account for differences in the degree of harm posed by different instances of fake news. To address these limitations, we design an Event-Level Evidence
Retrieval Framework (ELERF) and propose a Relation-Aware Evidence Graph Network (RAEGNet). ELERF retrieves external evidence based on the complete event semantics of a news item. RAEGNet constructs a directed graph that incorporates news-evidence stance relations and evidence-evidence interaction relations, and introduces a conditional-harm branch to jointly model authenticity and potential harm. Experimental results demonstrate that RAEGNet outperforms multiple baseline methods across all evaluated metrics on Weibo-21, Fakeddit, and our self-constructed SSS
dataset.
Read the full abstract ▾
All authors Wenbin Shen, Guoxuan Qin, Guangxu Yao, Baodong Wang, Yuanbo Rui, Zhongjie Ba, Zhichao Lian
Computation and Language
Momentum-Coupled Rubric Adaptation for Detailed Image Captioning
Zhenwen Ji, Lei Jin, Shanyong Wang et al. (+6 more)
Detailed image captioning requires accurate and comprehensive descriptions of fine-grained visual content, yet caption quality spans factual accuracy, information coverage, and clarity. Compared with conventional methods that rely mainly on high-quality supervision or holistic rewards, rubric-based reinforcement learning decomposes these requirements into explicit criteria and provides targeted, structured feedback. However, existing methods often use separate models for caption generation, rubric construction, and judging, which may lead to inconsistent interpretations across roles. Some dynamic rubric methods alternate updates between the caption policy and rubric generator while keeping the judge fixed, but staged optimization may still leave rubric construction and judging out of step with policy optimization. We propose MoCo Rubric, a two-stage framework that coordinates these roles. First, role-conditioned, shared-parameter multi-task supervised fine-tuning equips a single
vision–
language model to serve as the Caption Policy, Rubric Generator, and Rubric Judge. Then, the Generator constructs rubrics online from captions sampled by the current Policy, reference captions, and image evidence. The Judge provides rubric-based rewards, and only the Policy receives GRPO updates. As Policy updates change the candidates being evaluated, we use an exponential moving average of the Policy parameters to update one momentum model shared by the Generator and Judge. This gradual transfer lets both rubric roles track Policy updates without separate RL optimization while smoothing parameter changes that could disrupt their rubric capabilities under direct synchronization. Across five captioning
benchmarks, MoCo Rubric achieves an average pairwise win rate of 72.83%, the best mean rank in blind ranking, and the highest average score in caption-based question answering.
Read the full abstract ▾
All authors Zhenwen Ji, Lei Jin, Shanyong Wang, Jiaming Lu, Chengqiang Lu, Yi Wu, Yao Hu, Lizhen Cui, Yanyu Xu
Computation and Language · Cross-listed from cs.AI
Harness Evolution as Learning: Approximation, Generalization, and Optimization Limits of Self-Improving Personal Agents
Zeyu Gan, Zixuan Gong and Yong Liu
As the capabilities of large
language models (LLMs) continue to advance, increasing
attention is turning to how to translate their abilities into useful behavior. Personal
agents bring this question into everyday settings, where models are expected to serve individual users and continually adapt to their preferences. With the underlying model held fixed, such adaptation relies on harness engineering: designing and evolving the surrounding layer that manages context,
memory, tools, and execution. Despite rapid progress, the factors governing effective harness
evolution remain insufficiently understood. To narrow this gap, we investigate three central questions concerning harness architecture, harness scale, and self-
evolution algorithms through complementary empirical and theoretical analyses. Empirically, we introduce a preference-oriented
benchmark and systematically characterize the capabilities and limitations of personal
agents associated with these three dimensions. Theoretically, we formulate harness
evolution as a learning problem and explain these phenomena through approximation, generalization, and optimization errors. Analyses of reachable policies, capacity under finite interaction evidence, and biased update dynamics provide theoretical accounts of the observed phenomena. Together, these results offer a unified perspective on the limits of personalization through harness
evolution and inform future harness design.
Read the full abstract ▾