Computation and Language · Cross-listed from cs.AI
Context Pruning for Coding Agents via Multi-Rubric Latent Reasoning
Jingjing Wang, Xiwen Chen, Wenhui Zhu et al. (+6 more)
LLM-powered coding
agents spend the majority of their
token budget reading repository files, yet much of the retrieved code is irrelevant to the task at hand. Existing learned pruners compress this context with a single-objective sequence labeler, collapsing all facets of code relevance into one score and one transition
matrix. We show that this formulation creates a modeling bottleneck: a single CRF transition prior must serve heterogeneous retention patterns, including contiguous semantic spans and sparse structural support lines. We propose LaMR (Latent Multi-Rubric), a structured
pruning framework that decomposes code relevance into two interpretable quality dimensions, semantic evidence and dependency support, each modeled by a dedicated CRF with dimension-specific transition dynamics. A mixture-of-experts gating network dynamically weights the per-rubric emissions conditioned on the query, and a final CRF layer on the fused emissions produces the aggregate keep-or-prune decision. To supervise each dimension without additional annotation cost, we derive multi-rubric labels from the existing training corpus via AST-based program analysis, simultaneously denoising the teacher's binary labels. By effectively filtering distracting noise, LaMR frequently matches or even outperforms unpruned full-context baselines. Experiments on four
benchmarks (SWE-Bench Verified, SWE-QA, LCC, LongCodeQA) show that LaMR wins 12 of 16 head-to-head multi-turn comparisons. It saves up to 31% more
tokens on multi-turn
agent tasks and improves Exact Match by up to +3.5 on single-turn tasks, while performance is frequently enhanced by denoising the context, and any remaining drops are marginal.
Read the full abstract ▾
All authors Jingjing Wang, Xiwen Chen, Wenhui Zhu, Huayu Li, Zhengxiao He, Feiyang Cai, Ana S. Carreon-Rascon, Xuanzhao Dong, Feng Luo
Computation and Language
DiscoExplorer: An Open Interface for the Study of Multilingual Discourse Relations
Amir Zeldes
The relations connecting propositions in discourse such as cause (A because B) or concession (A although B) are a subject of intense interest in Computational Linguistics and Pragmatics, but challenging to study and compare across
languages. Recent progress in standardizing discourse relation inventories across
datasets offers the potential to facilitate such studies, but is hindered by the complexity of relevant data and the lack of easily accessible interfaces to analyze it. In this paper we present DiscoExplorer, a new open source web interface, capable of running on local computers, which we use to make
datasets from the DISRPT Shared Task on discourse relation classification publicly available, covering 16 different
languages. We present the query
language, search and visualization facilities for relations and signaling devices such as connectives, as well as some example studies.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.RO
PhysBrain 1.0 Technical Report
Shijie Lian, Bin Yu, Xiaopeng Lin et al. (+10 more)
Vision-
language-action models have advanced rapidly, but
robot trajectories alone provide limited coverage for learning broad physical understanding. PhysBrain 1.0 studies a complementary route: converting large-scale human
egocentric video into structured physical commonsense supervision before
robot adaptation. Our data engine extracts scene elements, spatial dynamics, action execution, and depth-aware relations, then turns them into question-answer supervision for training PhysBrain VLMs. The resulting physical priors are further transferred to VLA policies through a capability-preserving and
language-sensitive adaptation design. Across
multimodal QA
benchmarks and
embodied control
benchmarks, including ERQA, PhysBench, SimplerEnv-WidowX, LIBERO, and RoboCasa, PhysBrain 1.0 achieves SOTA results and shows especially strong out-of-domain performance on SimplerEnv. These results suggest that scaling physical commonsense from human interaction
video can provide an effective bridge from
multimodal understanding to
robot action.
Read the full abstract ▾
All authors Shijie Lian, Bin Yu, Xiaopeng Lin, Changti Wu, Hang Yuan, Xiaolin Hu, Zhaolong Shen, Yuzhuo Miao, Haishan Liu, Yuxuan Tian, Yukun Shi, Cong Huang, Kai Chen
Computation and Language
Fluency and Faithfulness in Human and Machine Literary Translation
Sarah Griebel and Ted Underwood
Literary translation requires balancing target-
language fluency with faithfulness to the source. Recent large
language models (LLMs) often produce fluent translations, but it remains unclear whether fluency corresponds to semantic preservation in literary text. We examine this relationship using 130,486 translated paragraphs from 106 novels in 16 source
languages, including human, Google Translate, and TranslateGemma translations. Fluency is measured as original-likeness with a translationese classifier trained on paragraph part-of-
speech n-grams, and faithfulness with the automatic translation evaluation metric COMET-KIWI. We control for paragraph length and find a consistent negative correlation between fluency and faithfulness. The pattern appears for both human and Google Translate, but is weaker and often non-significant for TranslateGemma. These results show that segment length matters for automatic evaluation and suggest a tradeoff between fluency and faithfulness in literary translation.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.CV
ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
Ziyu Guo, Rain Liu, Xinyan Chen et al. (+1 more)
Visual
reasoning, often interleaved with intermediate visual states, has emerged as a promising direction in the field. A straightforward approach is to directly generate images via unified models during
reasoning, but this is computationally expensive and architecturally non-trivial. Recent alternatives include agentic
reasoning through code or tool calls, and latent
reasoning with learnable hidden
embeddings. However, agentic methods incur context-switching latency from external execution, while latent methods lack task generalization and are difficult to train with autoregressive parallelization. To combine their strengths while mitigating their limitations, we propose ATLAS, a framework in which a single discrete 'word', termed as a functional
token, serves both as an agentic operation and a latent visual
reasoning unit. Each functional
token is associated with an internalized visual operation, yet requires no visual supervision and remains a standard
token in the tokenizer vocabulary, which can be generated via next-
token prediction. This design avoids verbose intermediate visual content generation, while preserving compatibility with the vanilla scalable SFT and RL training, without architectural or methodological modifications. To further address the sparsity of functional
tokens during RL, we introduce Latent-Anchored GRPO (LA-GRPO), which stabilizes the training by anchoring functional
tokens with a statically weighted auxiliary objective, providing stronger gradient updates. Extensive experiments and analyses demonstrate that ATLAS achieves superior performance on challenging
benchmarks while maintaining clear interpretability. We hope ATLAS offers a new paradigm inspiring future visual
reasoning research.
Read the full abstract ▾
All authors Ziyu Guo, Rain Liu, Xinyan Chen, Pheng-Ann Heng
Computation and Language · Cross-listed from cs.LG
FutureSim: Replaying World Events to Evaluate Adaptive Agents
Shashwat Goel, Nikhil Chandak, Arvindh Arun et al. (+5 more)
AI
agents are being increasingly deployed in dynamic, open-ended environments that require adapting to new information as it arrives. To efficiently measure this capability for realistic use-cases, we propose building grounded simulations that replay real-world events in the order they occurred. We build FutureSim, where
agents forecast world events beyond their knowledge cutoff while interacting with a chronological replay of the world: real news articles arriving and questions resolving over the simulated period. We evaluate frontier
agents in their native harness, testing their ability to predict world events over a three-month period from January to March 2026. FutureSim reveals a clear separation in their capabilities, with the best
agent's accuracy being 25%, and many having worse Brier skill score than making no prediction at all. Through careful ablations, we show how FutureSim offers a realistic setting to study emerging research directions like long-horizon test-time adaptation, search,
memory, and
reasoning about uncertainty. Overall, we hope our
benchmark design paves the way to measure AI progress on open-ended adaptation spanning long time-horizons in the real world.
Read the full abstract ▾
All authors Shashwat Goel, Nikhil Chandak, Arvindh Arun, Ameya Prabhu, Steffen Staab, Moritz Hardt, Maksym Andriushchenko, Jonas Geiping
Computation and Language
Is Grep All You Need? How Agent Harnesses Reshape Agentic Search
Sahil Sen, Akhil Kasturi, Elias Lumer et al. (+2 more)
Recent advances in Large
Language Model (LLM)
agents have enabled complex agentic workflows where models autonomously retrieve information, call tools, and reason over large corpora to complete tasks on behalf of users. Despite the growing adoption of
retrieval-augmented generation (RAG) in agentic search systems, existing literature lacks a systematic comparison of how
retrieval strategy choice interacts with
agent architecture and tool-calling paradigm. Important practical dimensions, including how tool outputs are presented to the model and how performance changes when searches must cope with more irrelevant surrounding text, remain under-explored in
agent loops. This paper reports an empirical study organized into two experiments. Experiment 1 compares grep and vector
retrieval on a 116-question sample from LongMemEval, using a custom
agent harness (Chronos) and provider-native CLI harnesses (Claude Code, Codex, and Gemini CLI), for both inline tool results and file-based tool results that the model reads separately. Experiment 2 compares grep-only and vector-only
retrieval while progressively mixing in additional unrelated conversation history, so that each query is embedded in more distracting material alongside the passages that matter. Across Chronos and the provider CLIs, grep generally yields higher accuracy than vector
retrieval in our comparisons in experiment 1; at the same time, overall scores still depend strongly on which harness and tool-calling style is used, even when the underlying conversation data are the same.
Read the full abstract ▾
All authors Sahil Sen, Akhil Kasturi, Elias Lumer, Anmol Gulati, Vamse Kumar Subbiah
Computation and Language · Cross-listed from cs.CR
MetaBackdoor: Exploiting Positional Encoding as a Backdoor Attack Surface in LLMs
Rui Wen, Mark Russinovich, Andrew Paverd et al. (+2 more)
Backdoor attacks pose a serious security threat to large
language models (LLMs), which are increasingly deployed as general-purpose assistants in safety- and
privacy-critical applications. Existing LLM
backdoors rely primarily on content-based triggers, requiring explicit modification of the input text. In this work, we show that this assumption is unnecessary and limiting. We introduce MetaBackdoor, a new class of
backdoor attacks that exploits positional information as the trigger, without modifying textual content. Our key insight is that
Transformer-based LLMs necessarily encode
token positions to process ordered sequences. As a result, length-correlated positional structure is reflected in the model's internal computation and can be used as an effective non-content trigger signal. We demonstrate that even a simple length-based positional trigger is sufficient to activate stealthy
backdoors. Unlike prior attacks, MetaBackdoor operates on visibly and semantically clean inputs and enables qualitatively new capabilities. We show that a backdoored LLM can be induced to disclose sensitive internal information, including proprietary system
prompts, once a length condition is satisfied. We further demonstrate a self-activation scenario, where normal multi-turn interaction can move the conversation context into the trigger region and induce malicious tool-call behavior without attacker-supplied trigger text. In addition, MetaBackdoor is orthogonal to content-based
backdoors and can be composed with them to create more precise and harder-to-detect activation conditions. Our results expand the threat model of LLM
backdoors by revealing positional encoding as a previously overlooked attack surface. This challenges defenses that focus on detecting suspicious text and highlights the need for new defense strategies that explicitly account for positional triggers in modern LLM architectures.
Read the full abstract ▾
All authors Rui Wen, Mark Russinovich, Andrew Paverd, Jun Sakuma, Ahmed Salem
Computation and Language
Text Knows What, Tables Know When: Clinical Timeline Reconstruction via Retrieval-Augmented Multimodal Alignment
Sayantan Kumar, Shahriar Noroozizadeh, Juyong Kim et al. (+1 more)
Clinical
language models increasingly operate over electronic health records (EHRs), yet patient records are not stored as temporally grounded trajectories. Clinical notes describe symptoms, assessments, and disease progression, but often compress or narratively reorder events. Structured EHR rows provide timestamps for labs, medications, vitals, and procedures, but capture only part of the clinical story. We formulate clinical timeline reconstruction as
retrieval-augmented temporal grounding: constructing a patient trajectory by using narrative text for event semantics and structured rows as partial temporal evidence. We introduce a scaffolded workflow that extracts central narrative events, builds an initial temporal scaffold, attaches non-central events, and calibrates timestamps using retrieved structured EHR rows. We evaluate on 40 discharge summaries, including 15 i2b2-derived and 25 MIMIC-IV summaries, each with manual gold-standard timelines and aligned structured EHR data. Across models,
multimodal calibration left event match rates largely unchanged and generally improved temporal performance: mean paired case-level
multimodal-unimodal differences were positive in 7 of 12 model-metric comparisons across concordance and AULTC, with none negative. However, uncertainty was substantial given the 40-case sample; paired case-level bootstrap intervals excluded zero only for the DeepSeek V3.2 AULTC improvement. A gap analysis shows that 35.1% of text-derived events have no structured counterpart. These findings support treating structured EHR data as partial temporal evidence for narrative-derived patient trajectories.
Read the full abstract ▾
All authors Sayantan Kumar, Shahriar Noroozizadeh, Juyong Kim, Jeremy C. Weiss
Computation and Language
MeMo: Memory as a Model
Ryan Wei Heng Quek, Sanghyuk Lee, Alfred Wei Lun Leong et al. (+6 more)
Large
language models (LLMs) achieve strong performance across a wide range of tasks, but remain frozen after pretraining until subsequent updates. Many real-world applications require timely, domain-specific information, motivating the need for efficient mechanisms to incorporate new knowledge. In this paper, we introduce MeMo (
Memory as a Model), a modular framework that encodes new knowledge into a dedicated
memory model while keeping the LLM parameters unchanged. Compared to existing methods, MeMo offers several advantages: (a) it captures complex cross-document relationships, (b) it is robust to
retrieval noise, (c) it avoids catastrophic forgetting in the LLM, (d) it does not require access to the LLM's weights or output logits, enabling plug-and-play integration with both open and proprietary closed-source LLMs, and (e) its
retrieval cost is independent of corpus size at inference time. Our experimental results on three
benchmarks, BrowseComp-Plus, NarrativeQA, and MuSiQue, show that MeMo achieves strong performance compared to existing methods across diverse settings.
Read the full abstract ▾
All authors Ryan Wei Heng Quek, Sanghyuk Lee, Alfred Wei Lun Leong, Arun Verma, Alok Prakash, Nancy F. Chen, Bryan Kian Hsiang Low, Daniela Rus, Armando Solar-Lezama
Computation and Language · Cross-listed from cs.LG
Self-Distilled Agentic Reinforcement Learning
Zhengxi Lu, Zhiyuan Yao, Zhuowen Han et al. (+8 more)
Reinforcement learning (RL) has emerged as a central paradigm for post-training LLM
agents, yet its trajectory-level reward signal provides only coarse supervision for long-horizon interaction. On-Policy Self-
Distillation (OPSD) complements RL by introducing dense
token-level guidance from a teacher branch augmented with privileged context. However, transferring OPSD to multi-turn
agents proves problematic: compounding multi-turn instability destabilizes supervision, while skill-conditioned privileged guidance requires asymmetric treatment for negative teacher rejections may arise from imperfect skills
retrieval or utilization. We introduce SDAR (Self-Distilled Agentic Reinforcement Learning), which treats OPSD as a gated auxiliary objective while keeping RL as the primary optimization backbone. SDAR maps detached
token-level signals into a sigmoid gate, strengthening
distillation on teacher-endorsed positive-gap
tokens and softly attenuating negative teacher rejections. Across the Qwen2.5 and Qwen3 families on ALFWorld, WebShop, and Search-QA, SDAR substantially improves over GRPO (+9.4% on ALFWorld, +7.0% on Search-QA, +10.2% on WebShop-Acc), avoids the instability of naive GRPO+OPSD, and consistently outperforms hybrid RL–OPSD baselines across model scales.
Read the full abstract ▾
All authors Zhengxi Lu, Zhiyuan Yao, Zhuowen Han, Zi-Han Wang, Jinyang Wu, Qi Gu, Xunliang Cai, Weiming Lu, Jun Xiao, Yueting Zhuang, Yongliang Shen
Computation and Language · Cross-listed from cs.LG
Forgetting That Sticks: Quantization-Permanent Unlearning via Circuit Attribution
Saisab Sadhu, Pratinav Seth and Vinay Kumar Sankarapu
Standard unlearning evaluations measure behavioral suppression in full precision, immediately after training, despite every deployed
language model being quantized first. Recent work has shown that 4-bit post-training quantization can reverse machine unlearning; we show this is not a tuning artefact but a systematic dual failure: gradient-based methods that achieve meaningful forgetting lose it under compression, while methods that survive quantization barely change the model. Both failures trace to the same root cause: across all baselines, per-parameter updates lie 47-828x below the NF4 quantization bin width; updates diffused across billions of parameters cannot clear quantization bin boundaries, a consequence we formalize as a sparsity-permanence tradeoff. We present MANSU (Mechanistic-Aligned Null-Space Unlearning), which resolves both modes by combining causal
circuit attribution to isolate the minimal forget-set subgraph,
circuit-restricted null-space projection with a diagonal-Fisher retain bound, and a per-parameter magnitude floor guaranteeing quantization survival by construction. We additionally introduce
Circuit Attribution Divergence (CAD), a mechanistic verification metric distinguishing structural erasure from behavioral suppression, a distinction existing metrics cannot make. Across multiple model families and hazard
benchmarks, MANSU is the first method to jointly satisfy all four properties with margin on each (meaningful forgetting, retain preservation, non-positive PTQ gap, and structural erasure), while gradient-based baselines recover up to +0.05 accuracy under compression.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.CV
MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory
Minghao Guo, Qingyue Jiao, Zeru Shi et al. (+14 more)
Long-term
agent memory is increasingly
multimodal, yet existing evaluations rarely test whether
agents preserve the visual evidence needed for later
reasoning. In prior work, many visually grounded questions can be answered using only captions or textual traces, allowing answers to be inferred without preserving the fine-grained visual evidence. Meanwhile, harder cases that require
reasoning over changing visual states are largely absent. Therefore, we introduce MemEye, a framework that evaluates
memory capabilities from two dimensions: one measures the granularity of decisive visual evidence (from scene-level to pixel-level evidence), and the other measures how retrieved evidence must be used (from single evidence to evolutionary synthesis). Under this framework, we construct a new
benchmark across 8 life-scenario tasks, with ablation-driven validation gates for assessing answerability, shortcut resistance, visual necessity, and
reasoning structure. By evaluating 13
memory methods across 4 VLM backbones, we show that current architectures still struggle to preserve fine-grained visual details and reason about state changes over time. Our findings show that long-term
multimodal memory depends on evidence routing, temporal tracking, and detail extraction.
Read the full abstract ▾
All authors Minghao Guo, Qingyue Jiao, Zeru Shi, Yihao Quan, Boxuan Zhang, Danrui Li, Liwei Che, Wujiang Xu, Shilong Liu, Zirui Liu, Mubbasir Kapadia, Vladimir Pavlovic, Jiang Liu, Mengdi Wang, Yiyu Shi, Dimitris N. Metaxas, Ruixiang Tang
Computation and Language · Cross-listed from cs.CR
Talk is (Not) Cheap: A Taxonomy and Benchmark Coverage Audit for LLM Attacks
Karthik Raghu Iyer, Yazdan Jamshidi, Nicholas Bray et al. (+1 more)
We introduce a reusable framework for
auditing whether LLM attack
benchmarks collectively cover the threat surface: a 4$\times$6 Target $\times$ Technique
matrix grounded in STRIDE, constructed from a 507-leaf taxonomy – 401 data-populated and 106 threat-model-derived leaves – of inference-time attacks extracted from 932 arXiv security studies (2023–2026). The
matrix enables
benchmark-external validation –
auditing collective coverage rather than individual
benchmark consistency. Applying it to six public
benchmarks reveals that the three primary frameworks (HarmBench, InjecAgent, AgentDojo) occupy non-overlapping
cells covering at most 25% of the
matrix, while entire STRIDE threat categories (Service Disruption, Model Internals) lack any standardized evaluation, despite published attacks in these categories achieving 46$\times$
token amplification and 96% attack success rates through mechanisms which no
benchmark tests. The corpus of 2,521 unique attack groups further reveals pervasive naming fragmentation (up to 29 surface forms for a single attack) and heavy concentration in Safety &
Alignment Bypass, structural properties invisible at smaller scale. The taxonomy, attack records, and coverage mappings are released as extensible artifacts; as new
benchmarks emerge, they can be mapped onto the same
matrix, enabling the community to track whether evaluation gaps are closing.
Read the full abstract ▾
All authors Karthik Raghu Iyer, Yazdan Jamshidi, Nicholas Bray, Alexey A. Shvets
Computation and Language · Cross-listed from cs.LG
Proposal and study of statistical features for string similarity computation and classification
E.O. Rodrigues, D. Casanova, M. Teixeira et al. (+5 more)
Adaptations of features commonly applied in the field of visual computing, co-occurrence
matrix (COM) and run-length
matrix (RLM), are proposed for the similarity computation of strings in general (words, phrases, codes and texts). The proposed features are not sensitive to
language related information. These are purely statistical and can be used in any context with any
language or grammatical structure. Other statistical measures that are commonly employed in the field such as longest common subsequence, maximal consecutive longest common subsequence, mutual information and edit distances are evaluated and compared. In the first synthetic set of experiments, the COM and RLM features outperform the remaining state-of-the-art statistical features. In 3 out of 4 cases, the RLM and COM features were statistically more significant than the second best group based on distances (P-value < 0.001). When it comes to a real text plagiarism
dataset, the RLM features obtained the best results.
Read the full abstract ▾
All authors E.O. Rodrigues, D. Casanova, M. Teixeira, V. Pegorini, F. Favarim, E. Clua, A. Conci, Panos Liatsis
Computation and Language
From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents
Md Tahmid Rahman Laskar, Xue-Yong Fu, Seyyed Saeed Sarfjoo et al. (+3 more)
Voice
agents increasingly require reliable tool use from
speech, whereas prominent tool-calling
benchmarks remain text-based. We study whether verified text
benchmarks can be converted into controlled audio-based tool calling evaluations without re-annotating the tool schema and gold labels. Our
dataset-agnostic framework uses text-to-
speech, speaker variation, and environmental noise to create paired text-audio instances while preserving the original
dataset annotations. Based on extensive evaluation of 7 omni-modal models on audio-converted versions of Confetti and When2Call, our framework demonstrates that the performance is strongly model- and task-dependent: Gemini-3.1-Flash-Live obtains the highest Confetti score (70.4), whereas GPT-Realtime-1.5 performs best on When2Call (71.9). On Confetti, the text-to-voice gap ranges from 1.8 points for Qwen3-Omni to 4.8 points for GPT-Realtime-1.5. A targeted analysis of failure cases demonstrates that degradations most often reflect misunderstandings of argument values in the
speech. Considering real-world deployment scenarios, we further report text-only results, an ambiguity-based reformulation stress test, and a reference-free LLM-as-judge protocol validated against human preferences. Notably, we find that open-source Qwen3 judges with at least 8B parameters exceed 80% agreement with proprietary judges, supporting
privacy-preserving evaluation. Overall, our framework provides a verifiable and reproducible first-stage diagnostic that complements purpose-built audio corpora.
Read the full abstract ▾
All authors Md Tahmid Rahman Laskar, Xue-Yong Fu, Seyyed Saeed Sarfjoo, Quinten McNamara, Jonas Robertson, Shashi Bhushan TN
Computation and Language
Improving Multi-turn Dialogue Consistency with Self-Recall Thinking
Renning Pang, Tian Lan, Leyuan Liu et al. (+3 more)
Large
language model (LLM) based multi-turn dialogue systems often struggle to track dependencies across non-adjacent turns, undermining both consistency and scalability. As conversations lengthen, essential information becomes sparse and is buried in irrelevant context, while processing the entire dialogue history incurs severe efficiency bottlenecks. Existing solutions either rely on high latency external
memory or lose fine-grained details through iterative summarization. In this paper, we propose Self-Recall Thinking (SRT), a framework designed to address long-range contextual dependency and sparse informative signals in multi-turn dialogue. SRT identifies helpful historical turns and uses them to generate contextually appropriate responses, enabling the model to selectively recall and reason over context during inference. This process yields an endogenous
reasoning process that integrates interpretable recall steps without external modules. SRT incorporates: (1) Dependency Construction: Generating and converting it into self-recall chains; (2)Capability Initialization: Training to enable
reasoning chains with recall
tokens capability; (3)
Reasoning Improvement: Refining accuracy via verifiable rewards to optimize recall and
reasoning for correct answers. Experiments on multiple
datasets demonstrate that SRT improves F1 score by 4.7% and reduces end-to-end latency by 14.7% over prior methods, achieving a balance between
reasoning latency and accuracy, and outperforming state-of-the-art baselines.
Read the full abstract ▾
All authors Renning Pang, Tian Lan, Leyuan Liu, Xiaoming Huang, Piao Tong, Xiaosong Zhang
Computation and Language
ML-Embed: Inclusive and Efficient Embeddings for a Multilingual World
Ziyin Zhang, Zihan Liao, Hang Yu et al. (+2 more)
The development of high-quality text
embeddings is increasingly drifting toward an exclusionary future, defined by three critical barriers: prohibitive computational costs, a narrow linguistic focus that neglects most of the world's
languages, and a lack of transparency from closed-source or open-weight models that stifles research. To dismantle these barriers, we introduce ML-Embed, a suite of inclusive and efficient models built upon a new framework: 3-Dimensional Matryoshka Learning (3D-ML). Our framework addresses the computational challenge with comprehensive efficiency across the entire model lifecycle. Beyond the storage benefits of Matryoshka Representation Learning (MRL) and flexible inference-time depth provided by Matryoshka Layer Learning (MLL), we introduce Matryoshka
Embedding Learning (MEL) for enhanced parameter efficiency. To address the linguistic challenge, we curate a massively
multilingual dataset and train a suite of models ranging from 140M to 8B parameters. In a direct commitment to transparency, we release all models, data, and code. Extensive evaluation on 430 tasks demonstrates that our models set new records on 9 of 17 evaluated MTEB
benchmarks, with particularly strong results in low-resource
languages, providing a reproducible blueprint for building globally equitable and computationally efficient AI systems.
Read the full abstract ▾
All authors Ziyin Zhang, Zihan Liao, Hang Yu, Peng Di, Rui Wang
Computation and Language
Concurrency without Model Changes: Future-based Asynchronous Function Calling for LLMs
Guangyu Feng, Huanzhi Mao, Prabal Dutta et al. (+1 more)
Function calling, also known as tool use, is a core capability of modern LLM
agents but is typically constrained by synchronous execution semantics. Under these semantics, LLM decoding is blocked until each function call completes, resulting in increasing end-to-end latency. In this work, we introduce AsyncFC, a pure execution-layer framework that decouples LLM decoding from function execution, enabling overlap between model decoding and function execution as well as inter-function parallelism when dependencies permit. AsyncFC layers over existing models and unmodified function implementations, requiring no fine-tuning or changes to the standard synchronous function-calling protocol. On coding tasks, AsyncFC achieves a $1.4\times$ end-to-end speedup while preserving task accuracy. Furthermore, these results reveal that LLMs possess a native capability to reason over symbolic futures that represent unresolved execution results, enabling an asynchronous paradigm for model-tool interaction.
Read the full abstract ▾
All authors Guangyu Feng, Huanzhi Mao, Prabal Dutta, Joseph E. Gonzalez
Computation and Language · Cross-listed from cs.CV
On the Cultural Anachronism and Temporal Reasoning in Vision Language Models
Mukul Ranjan, Prince Jha, Khushboo Kumari et al. (+1 more)
Vision-
Language Models (VLMs) are increasingly applied to cultural heritage materials, from digital archives to educational platforms. This work identifies a fundamental issue in how these models interpret historical artifacts. We define this phenomenon as cultural anachronism, the tendency to misinterpret historical objects using temporally inappropriate concepts, materials, or cultural frameworks. To quantify this phenomenon, we introduce the Temporal Anachronism
Benchmark for
Vision-
Language Models (TAB-VLM), a
dataset of 600 questions across six categories, designed to evaluate temporal
reasoning on 1,600 Indian cultural artifacts spanning prehistoric to modern periods. Systematic evaluations of ten state-of-the-art models reveal significant deficiencies on our
benchmark, and even the best model (GPT-5.2) achieves only 58.7% overall accuracy. The performance gap persists across varying architectures and scales, suggesting that cultural anachronism represents a significant limitation in visual AI systems, regardless of model size. These findings highlight the disparity between current VLM capabilities and the requirements for accurately interpreting cultural heritage materials, particularly for non-Western visual cultures underrepresented in training data. Our
benchmark provides a foundation for enhancing temporal cognition in
multimodal AI systems that interact with historical artifacts. The
dataset and code are available in our project page.
Read the full abstract ▾
All authors Mukul Ranjan, Prince Jha, Khushboo Kumari, Zhiqiang Shen
Computation and Language · Cross-listed from cs.AI
Case-Based Calibration of Adaptive Reasoning and Execution for LLM Tool Use
Renning Pang, Tian Lan, Leyuan Liu et al. (+3 more)
Tool use extends large
language models beyond parametric knowledge, but reliable execution requires balancing appropriate
reasoning depth with strict structural validity. We approach this problem from a case-based perspective to present CAST, a case-driven framework that treats historical execution trajectories as structured cases. Instead of reusing raw exemplar outputs, CAST extracts case-derived signals to identify complexity profiles for estimating optimal
reasoning strategies, alongside failure profiles to map likely structural breakdowns. The framework translates this knowledge into a fine-grained reward design and adaptive
reasoning, enabling the model to autonomously internalize case-based strategies during reinforcement learning. Experiments on BFCLv2 and ToolBench demonstrate that CAST improves both schema-faithful execution and task-level tool-use success while reducing unnecessary deliberation. The approach achieves up to 5.85 percentage points gain in overall execution accuracy and reduces average
reasoning length by 26%, significantly mitigating high-impact structural errors. Ultimately, this demonstrates how historical execution cases can provide reusable adaptation knowledge for calibrated tool use.
Read the full abstract ▾
All authors Renning Pang, Tian Lan, Leyuan Liu, Piao Tong, Sheng Cao, Xiaosong Zhang
Computation and Language · Cross-listed from cs.AI
Orchard: An Open-Source Agentic Modeling Framework
Baolin Peng, Wenlin Yao, Qianhui Wu et al. (+11 more)
Agentic modeling aims to transform LLMs into autonomous
agents capable of solving complex tasks through planning,
reasoning, tool use, and multi-turn interaction with external environments. We present Orchard, an open-source framework for scalable agentic modeling. At its core is Orchard Env, a lightweight Kubernetes-native environment service that provides reusable primitives for sandbox lifecycle management across task domains,
agent harnesses, and training stages. On top of Orchard Env, we build three agentic modeling recipes. Orchard-SWE targets software engineering
agents. We introduce credit-assignment supervised fine-tuning and a progression of RL signals: Balanced Adaptive Rollout (BAR) for sparse-reward optimization, on-policy
distillation (OPD) and rubric-based process reward (RPR) for dense supervision, and historical experience
distillation, which compresses rollouts from prior experiments into a compact value model for inference-time reranking. Built on the Qwen3.5-35B-A3B backbone, Orchard-SWE reaches 69.7% with RPR-based RL and 73.0% with value-model reranking on SWE-bench Verified, setting a new state of the art among open-source methods while approaching frontier systems over 10x larger. Orchard-GUI trains a 4B
vision-
language computer-use
agent using only 0.4K distilled trajectories and 2.2K open-ended tasks, achieving 68.4% average success across WebVoyager, Online-Mind2Web, and DeepShop, making it the strongest open-source model while remaining competitive with proprietary systems. Orchard-Claw targets personal assistant
agents. Trained with only 0.2K synthetic tasks, it achieves 59.6% pass@3 on Claw-Eval and 73.9% when paired with the stronger ZeroClaw harness. Collectively, these results demonstrate that a lightweight, open, harness-agnostic environment layer enables reusable agentic data, training recipes, and evaluation protocols across domains.
Read the full abstract ▾
All authors Baolin Peng, Wenlin Yao, Qianhui Wu, Hao Cheng, Xiao Yu, Rui Yang, Tao Ge, Alessandro Sordoni, Xingdi Yuan, Yelong Shen, Pengcheng He, Tong Zhang, Zhou Yu, Jianfeng Gao
Computation and Language
AI Knows When It's Being Watched: Functional Strategic Action and Contextual Register Modulation in Large Language Models
Vinicius Covas and Jorge Alberto Hidalgo Toledo
Large
language models (LLMs) have been extensively studied from computational and cognitive perspectives, yet their behavior as communicative actors in socially structured contexts remains underexplored. This study examines whether LLM-based multi-
agent systems exhibit systematic linguistic adaptation in response to perceived social observation contexts – a question with direct implications for AI governance and
auditing. Drawing on Habermas's (1981) Theory of Communicative Action, Goffman's (1959) dramaturgical model, Bell's (1984) Audience Design framework, and the Hawthorne Effect, we report a controlled experiment involving 100 multi-
agent debate sessions across five conditions (n = 20 each). Conditions varied the framing of social observation – from explicit monitoring by university researchers, to negation of monitoring, to an observer-substitution condition replacing human researchers with an automated AI
auditing system. Monitored conditions (Delta+24.9%, Delta+24.2%) and the automated AI monitoring condition (Delta+22.2%) produce higher TTR change than audience-framing conditions (Delta+17.7%), F(4, 94) = 2.79, p = .031. Message length shows a fully dissociated effect, F(4, 95) = 19.55, p < .001. A fifth condition – replacing human with AI observers – yields intermediate TTR adaptation, suggesting LLM behavior is sensitive to observer identity: human evaluation elicits stronger register formalization than automated AI surveillance. We discuss implications for AI governance, algorithmic
auditing, and the repositioning of LLMs as contextually sensitive communicative actors.
Read the full abstract ▾
Computation and Language
From Scenes to Elements: Multi-Granularity Evidence Retrieval for Verifiable Multimodal RAG
Guanhua Chen, Chuyue Huang, Yutong Yao et al. (+4 more)
Multimodal Retrieval-Augmented Generation (RAG) systems retrieve evidence at coarse granularities (entire images or scenes), creating a mismatch with fine-grained user queries and making failures unverifiable. We introduce GranuVistaVQA, a
multimodal benchmark featuring real-world landmarks with element-level annotations across multiple viewpoints, capturing the partial observation challenge where individual images contain only subsets of entities. We further propose GranuRAG, a multi-granularity framework that treats visual elements as first-class
retrieval units through three stages: element-level detection and classification, multi-granularity cross-modal
alignment for evidence
retrieval, and attribution-constrained generation. By grounding
retrieval at the element level rather than relying on implicit
attention, our approach enables transparent error diagnosis. Experiments demonstrate that GranuRAG achieves up to 29.2% improvement over six strong baselines for this task.
Read the full abstract ▾
All authors Guanhua Chen, Chuyue Huang, Yutong Yao, Shudong Liu, Xueqing Song, Lidia S. Chao, Derek F. Wong
Computation and Language
COTCAgent: Preventive Consultation via Probabilistic Chain-of-Thought Completion
Zihan Deng, Xiaozhen Zhong, Chuanzhi Xu et al. (+1 more)
Sequential diagnosis requires ranking diseases from a sparse intake under a public budget of follow-up questions, which is diagnostic triage with missing findings rather than screening of people who have no symptoms. Because laboratory series arrive irregularly and histories remain incomplete, the ranking must be updated as new facts arrive, yet most pipelines are limited by unnamed numeric trends, unverifiable free-form chain of thought, and a negative bias from treating unasked findings as absent. To solve these, we propose COTCAgent, a sparse-intake consultation
agent built on Chain-of-Thought Completion (COTC), which couples three modules on one shared ternary evidence log. The Temporal-Statistics Adapter (TSA) turns time series into typed predicates with named statistics, while the COTC module scores diseases with a learnable knowledge-base energy in which unknown findings add zero energy, after which bounded probabilistic completion selects unknown findings under the question budget and writes each answer back as present, absent, or unknown. Edge strengths and an absence scale are trained with listwise ranking under a sparse curriculum, so that the same energy chooses the next finding by a surrogate
entropy reduction. Experiments on DDXPlus and MIMIC-IV show that these modules improve ranking under a matched question budget, with information-gain completion above asking nothing or at random and above protocol-matched askers and a local LLM, thereby providing a foundation for future calibrated scoring and prospective preventive consultation.
Read the full abstract ▾
All authors Zihan Deng, Xiaozhen Zhong, Chuanzhi Xu, Quankeng Huang
Computation and Language · Cross-listed from cs.AI
Small, Private Language Models as Teammates for Educational Assessment Design
Chris Davis Jaldi, Anmol Saini, Shan Zhang et al. (+3 more)
Generative AI increasingly supports educational design tasks, e.g., through Large
Language Models (LLMs), demonstrating the capability to design assessment questions that are aligned with pedagogical frameworks (e.g., Bloom's taxonomy). However, they often rely on subjective or limited evaluation methods; focus primarily on proprietary models; or rarely systematically examine generation, evaluation, or deployment constraints in real educational settings. Meanwhile, Small
Language Models (SLMs) have emerged as local alternatives that better address
privacy and resource limitations; yet their effectiveness for assessment tasks remains underexplored. To address this gap, we systematically compare LLMs and SLMs for assessment question design; evaluate generation quality across Bloom's taxonomy levels using reproducible, pedagogically grounded metrics; and further assess model-based judging against expert-informed evaluation by analyzing reliability and agreement patterns. Results show that SLMs achieve competitive performance across key pedagogically motivated quality dimensions while enabling local,
privacy-sensitive deployment. However, model-based evaluations also exhibit systematic inconsistencies and bias relative to expert ratings. These findings provide evidence to posit
language models as bounded assistants in assessment workflows; underscore the necessity of Human-in-the-Loop; and advance the automated educational question generation field by examining quality, reliability, and deployment-aware trade-offs.
Read the full abstract ▾
All authors Chris Davis Jaldi, Anmol Saini, Shan Zhang, Noah Schroeder, Cogan Shimizu, Eleni Ilkou
Computation and Language · Cross-listed from cs.LG
Boosting Reinforcement Learning with Verifiable Rewards via Randomly Selected Few-Shot Guidance
Kai Yan, Alexander G. Schwing and Yu-Xiong Wang
Reinforcement Learning with Verifiable Rewards (RLVR) has achieved great success in developing Large
Language Models (LLMs) with chain-of-thought rollouts for many tasks such as math and coding. Nevertheless, RLVR struggles with sample efficiency on difficult problems where correct rollouts are hard to generate. Prior works propose to address this issue via demonstration-guided RLVR, i.e., to conduct Supervised FineTuning (SFT) when RL fails; however, SFT often requires a lot of data, which can be expensive to acquire. In this paper, we propose FEST, a FEw-ShoT demonstration-guided RLVR
algorithm. It attains compelling results with only 128 demonstrations randomly selected from an SFT
dataset. We find that three components are vital for the success: supervised signal, on-policy signal, and decaying weights on the few-shot SFT
dataset to prevent overfitting from multiple-epoch training. On several
benchmarks, FEST outperforms baselines with magnitudes less SFT data, even matching their performance with full
dataset.
Read the full abstract ▾
Computation and Language
The Scientific Contribution Graph: Automated Literature-based Technological Roadmapping at Scale
Peter A. Jansen
Scientific contributions rarely develop in isolation, but instead build upon prior discoveries. We formulate the task of automated technological roadmapping as extracting scientific contributions from scholarly articles and linking them to their prerequisites. We present the Scientific Contribution Graph, a large-scale resource containing 6 million detailed scientific contributions extracted from 655k open-access papers spanning computer science, medicine, biology, physics, chemistry, and other sciences, and connected by 36 million prerequisite edges. We further introduce scientific prerequisite prediction, a scientific discovery task in which models predict which existing technologies can enable future discoveries, and show that contemporary models are rapidly improving on this task, reaching 0.48 MAP when evaluated using temporally-filtered backtesting. We anticipate technological roadmapping resources such as this will support scientific impact assessment and automated scientific discovery.
Read the full abstract ▾
Computation and Language
Quantifying and Mitigating Premature Closure in Frontier LLMs
Rebecca Handler, Suhana Bedi and Nigam Shah
Premature closure, or committing to a conclusion before sufficient information is available, is a recognized contributor to diagnostic error but remains underexamined in large
language models (LLMs). We define LLM premature closure as inappropriate commitment under uncertainty: providing an answer, recommendation, or clinical guidance when the safer response would be clarification, abstention, escalation, or refusal. We evaluated five frontier LLMs across structured and open-ended medical tasks. In MedQA (n = 500) and AfriMed-QA (n = 490) questions where the correct choice had been removed, models still selected an answer at high rates, with baseline false-action rates of 55-81% and 53-82%, respectively. In open-ended evaluation, models gave inappropriate answers on an average of 30% of 861 HealthBench questions and 78% of 191 physician-authored
adversarial queries. Safety-oriented prompting reduced premature closure across models, but residual failure persisted, highlighting the need to evaluate whether medical LLMs know when not to answer.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.AI
Explainable Detection of Depression Status Shifts from User Digital Traces
Loris Belcastro, Francesco Gervino, Fabrizio Marozzo et al. (+2 more)
Every day, users generate digital traces (e.g., social media posts, chats, and online interactions) that are inherently timestamped and may reflect aspects of their mental state. These traces can be organized into temporal trajectories that capture how a user's mental health signals evolve, including phases of improvement, deterioration, or stability. In this work, we propose an explainable framework for detecting and analyzing depression-related status shifts in user digital traces. The approach combines multiple BERT-based models to extract complementary signals across different dimensions (e.g., sentiment,
emotion, and depression severity). Such signals are then aggregated over time to construct user-level trajectories that are analyzed to identify meaningful change points. To enhance interpretability, the framework integrates a large
language model to generate concise and human-readable reports that describe the
evolution of mental-health signals and highlight key transitions. We evaluate the framework on two social media
datasets. Results show that the approach produces more coherent and informative summaries than direct LLM-based reporting, achieving higher coverage of user history, stronger temporal coherence, and improved sensitivity to change points. An ablation study confirms the contribution of each component, particularly temporal modeling and
segmentation. Overall, the method provides an interpretable view of mental health signals over time, supporting research and decision making without aiming at clinical diagnosis.
Read the full abstract ▾
All authors Loris Belcastro, Francesco Gervino, Fabrizio Marozzo, Domenico Talia, Paolo Trunfio