Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Computation and Language

cs.CL Watch Computation and Language

Tuesday 3 February

Computation and Language

On the Credibility of Evaluating LLMs using Survey Questions

Jindřich Libovický

Recent studies evaluate the value orientation of large language models (LLMs) using adapted social surveys, typically by prompting models with survey questions and comparing their responses to average human responses. This paper identifies limitations in this methodology that, depending on the exact setup, can lead to both underestimating and overestimating the similarity of value orientation. Using the World Value Survey in three languages across five countries, we demonstrate that prompting methods (direct vs. chain-of-thought) and decoding strategies (greedy vs. sampling) significantly affect results. To assess the interaction between answers, we introduce a novel metric, self-correlation distance. This metric measures whether LLMs maintain consistent relationships between answers across different questions, as humans do. This indicates that even a high average agreement with human data, when considering LLM responses independently, does not guarantee structural alignment in responses. Additionally, we reveal a weak correlation between two common evaluation metrics, mean-squared distance and KL divergence, which assume that survey answers are independent of each other. For future research, we recommend CoT prompting, sampling-based decoding with dozens of samples, and robust analysis using multiple metrics, including self-correlation distance.
Read the full abstract

Computation and Language · Cross-listed from cs.HC

Chaplains' Reflections on the Design and Usage of AI for Conversational Care

Joel Wester, Samuel Rhys Cox, Henning Pohl et al. (+1 more)

Despite growing recognition that responsible AI requires domain knowledge, current work on conversational AI primarily draws on clinical expertise that prioritises diagnosis and intervention. However, much of everyday emotional support needs occur in non-clinical contexts, and therefore requires different conversational approaches. We examine how chaplains, who guide individuals through personal crises, grief, and reflection, perceive and engage with conversational AI. We recruited eighteen chaplains to build AI chatbots. While some chaplains viewed chatbots with cautious optimism, the majority expressed limitations of chatbots' ability to support everyday well-being. Our analysis reveals how chaplains perceive their pastoral care duties and areas where AI chatbots fall short, along the themes of Listening, Connecting, Carrying, and Wanting. These themes resonate with the idea of attunement, recently highlighted as a relational lens for understanding the delicate experiences care technologies provide. This perspective informs chatbot design aimed at supporting well-being in non-clinical contexts.
Read the full abstract

All authors Joel Wester, Samuel Rhys Cox, Henning Pohl, Niels van Berkel

Computation and Language

Transformers perform adaptive partial pooling

Vsevolod Kapatsinski

Any language model must decide what to say in novel contexts based on information from similar contexts. But what about contexts that are not novel but merely infrequent? In hierarchical regression, the model's predictions for behavior in a context are affected by observations from similar contexts to the extent that 1) the current context is infrequent and 2) different contexts behave similarly. This is called adaptive partial pooling. This paper shows that next-word predictions of a transformer (GPT2) are affected by observations from outside the current context, but this pooling reduces with more training. Pooling is affected by context frequency, context number (type frequency) and context variability in a qualitatively similar way to hierarchical regression. However, there is a "sweet spot" in training at which the transformer best matches the behavior of hierarchical regression. This is the point at which the effect of context frequency on pooling is at maximum.
Read the full abstract

Computation and Language

Likelihood-Based Reward Designs for General LLM Reasoning

Ariel Kwiatkowski, Natasha Butt, Ismail Labiad et al. (+2 more)

Fine-tuning large language models (LLMs) on reasoning benchmarks via reinforcement learning requires a specific reward function, often binary, for each benchmark. This comes with two potential limitations: the need to design the reward, and the potentially sparse nature of binary rewards. Here, we systematically investigate rewards derived from the probability or log-probability of emitting the reference answer (or any other prompt continuation present in the data), which have the advantage of not relying on specific verifiers and being available at scale. Several recent works have advocated for the use of similar rewards (e.g., VeriFree, JEPO, RLPR, NOVER). We systematically compare variants of likelihood-based rewards with standard baselines, testing performance both on standard mathematical reasoning benchmarks, and on long-form answers where no external verifier is available. We find that using the log-probability of the reference answer as the reward for chain-of-thought (CoT) learning is the only option that performs well in all setups. This reward is also consistent with the next-token log-likelihood loss used during pretraining. In verifiable settings, log-probability rewards bring comparable or better success rates than reinforcing with standard binary rewards, and yield much better perplexity. In non-verifiable settings, they perform on par with SFT. On the other hand, methods based on probability, such as VeriFree, flatline on non-verifiable settings due to vanishing probabilities of getting the correct answer. Overall, this establishes log-probability rewards as a viable method for CoT fine-tuning, bridging the short, verifiable and long, non-verifiable answer settings.
Read the full abstract

All authors Ariel Kwiatkowski, Natasha Butt, Ismail Labiad, Julia Kempe, Yann Ollivier

Computation and Language

Automatic Classification of Pedagogical Materials against CS Curriculum Guidelines

Erik Saule, Kalpathi Subramanian and Razvan Bunescu

Professional societies often publish curriculum guidelines to help programs align their content to international standards. In Computer Science, the primary standard is published by ACM and IEEE and provide detailed guidelines for what should be and could be included in a Computer Science program. While very helpful, it remains difficult for program administrators to assess how much of the guidelines is being covered by a CS program. This is in particular due to the extensiveness of the guidelines, containing thousands of individual items. As such, it is time consuming and cognitively demanding to audit every course to confidently mark everything that is actually being covered. Our preliminary work indicated that it takes about a day of work per course. In this work, we propose using Natural Language Processing techniques to accelerate the process. We explore two kinds of techniques, the first relying on traditional tools for parsing, tagging, and embeddings, while the second leverages the power of Large Language Models. We evaluate the application of these techniques to classify a corpus of pedagogical materials and show that we can meaningfully classify documents automatically.
Read the full abstract

Computation and Language

Linguistic Blind Spots in Clinical Decision Extraction

Mohamed Elgaar and Hadi Amiri

Extracting medical decisions from clinical notes is a key step for clinical decision support and patient-facing care summaries. We study how the linguistic characteristics of clinical decisions vary across decision categories and whether these differences explain extraction failures. Using MedDec discharge summaries annotated with decision categories from the Decision Identification and Classification Taxonomy for Use in Medicine (DICTUM), we compute seven linguistic indices for each decision span and analyze span-level extraction recall of a standard transformer model. We find clear category-specific signatures: drug-related and problem-defining decisions are entity-dense and telegraphic, whereas advice and precaution decisions contain more narrative, with higher stopword and pronoun proportions and more frequent hedging and negation cues. On the validation split, exact-match recall is 48%, with large gaps across linguistic strata: recall drops from 58% to 24% from the lowest to highest stopword-proportion bins, and spans containing hedging or negation cues are less likely to be recovered. Under a relaxed overlap-based match criterion, recall increases to 71%, indicating that many errors are span boundary disagreements rather than complete misses. Overall, narrative-style spans–common in advice and precaution decisions–are a consistent blind spot under exact matching, suggesting that downstream systems should incorporate boundary-tolerant evaluation and extraction strategies for clinical decisions.
Read the full abstract

Computation and Language · Cross-listed from cs.CV

SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?

Azmine Toushik Wasi, Wahid Faisal, Abdur Rahman et al. (+12 more)

Spatial reasoning is a fundamental aspect of human cognition, yet it remains a major challenge for contemporary vision-language models (VLMs). Prior work largely relied on synthetic or LLM-generated environments with limited task designs and puzzle-like setups, failing to capture the real-world complexity, visual noise, and diverse spatial relationships that VLMs encounter. To address this, we introduce SpatiaLab, a comprehensive benchmark for evaluating VLMs' spatial reasoning in realistic, unconstrained contexts. SpatiaLab comprises 1,400 visual question-answer pairs across six major categories: Relative Positioning, Depth & Occlusion, Orientation, Size & Scale, Spatial Navigation, and 3D Geometry, each with five subcategories, yielding 30 distinct task types. Each subcategory contains at least 25 questions, and each main category includes at least 200 questions, supporting both multiple-choice and open-ended evaluation. Experiments across diverse state-of-the-art VLMs, including open- and closed-source models, reasoning-focused, and specialized spatial reasoning models, reveal a substantial gap in spatial reasoning capabilities compared with humans. In the multiple-choice setup, InternVL3.5-72B achieves 54.93% accuracy versus 87.57% for humans. In the open-ended setting, all models show a performance drop of around 10-25%, with GPT-5-mini scoring highest at 40.93% versus 64.93% for humans. These results highlight key limitations in handling complex spatial relationships, depth perception, navigation, and 3D geometry. By providing a diverse, real-world evaluation framework, SpatiaLab exposes critical challenges and opportunities for advancing VLMs' spatial reasoning, offering a benchmark to guide future research toward robust, human-aligned spatial understanding. SpatiaLab is available at: https://spatialab-reasoning.github.io/.
Read the full abstract

All authors Azmine Toushik Wasi, Wahid Faisal, Abdur Rahman, Mahfuz Ahmed Anik, Munem Shahriar, Mohsin Mahmud Topu, Sadia Tasnim Meem, Rahatun Nesa Priti, Sabrina Afroz Mitu, Md. Iqramul Hoque, Shahriyar Zaman Ridoy, Mohammed Eunus Ali, Majd Hawasly, Mohammad Raza, Md Rizwan Parvez

Computation and Language

Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probing

Tong Zheng, Chengsong Huang, Runpeng Dai et al. (+9 more)

Parallel thinking has emerged as a promising paradigm for reasoning, yet it imposes significant computational burdens. Existing efficiency methods primarily rely on local, per-trajectory signals and lack principled mechanisms to exploit global dynamics across parallel branches. We introduce 2D probing, an interface that exposes the width-depth dynamics of parallel thinking by periodically eliciting intermediate answers from all branches. Our analysis reveals three key insights: non-monotonic scaling across width-depth allocations, heterogeneous reasoning branch lengths, and early stabilization of global consensus. Guided by these insights, we introduce $\textbf{{Parallel-Probe}}$, a training-free controller designed to optimize online parallel thinking. Parallel-Probe employs consensus-based early stopping to regulate reasoning depth and deviation-based branch pruning to dynamically adjust width. Extensive experiments across three benchmarks and multiple models demonstrate that Parallel-Probe establishes a superior Pareto frontier for test-time scaling. Compared to standard majority voting, it reduces sequential tokens by up to $\textbf{35.8}$% and total token cost by over $\textbf{25.8}$% while maintaining competitive accuracy.
Read the full abstract

All authors Tong Zheng, Chengsong Huang, Runpeng Dai, Yun He, Rui Liu, Xin Ni, Huiwen Bao, Kaishen Wang, Hongtu Zhu, Jiaxin Huang, Furong Huang, Heng Huang

Computation and Language

Accelerating Scientific Research with Gemini: Case Studies and Common Techniques

David P. Woodruff, Vincent Cohen-Addad, Lalit Jain et al. (+33 more)

Recent advances in large language models (LLMs) have opened new avenues for accelerating scientific research. While models are increasingly capable of assisting with routine tasks, their ability to contribute to novel, expert-level mathematical discovery is less understood. We present a collection of case studies demonstrating how researchers have successfully collaborated with advanced AI models, specifically Google's Gemini-based models (in particular Gemini Deep Think and its advanced variants), to solve open problems, refute conjectures, and generate new proofs across diverse areas in theoretical computer science, as well as other areas such as economics, optimization, and physics. Based on these experiences, we extract common techniques for effective human-AI collaboration in theoretical research, such as iterative refinement, problem decomposition, and cross-disciplinary knowledge transfer. While the majority of our results stem from this interactive, conversational methodology, we also highlight specific instances that push beyond standard chat interfaces. These include deploying the model as a rigorous adversarial reviewer to detect subtle flaws in existing proofs, and embedding it within a "neuro-symbolic" loop that autonomously writes and executes code to verify complex derivations. Together, these examples highlight the potential of AI not just as a tool for automation, but as a versatile, genuine partner in the creative process of scientific discovery.
Read the full abstract

All authors David P. Woodruff, Vincent Cohen-Addad, Lalit Jain, Jieming Mao, Song Zuo, MohammadHossein Bateni, Simina Branzei, Michael P. Brenner, Lin Chen, Ying Feng, Lance Fortnow, Gang Fu, Ziyi Guan, Zahra Hadizadeh, Mohammad T. Hajiaghayi, Mahdi JafariRaviz, Adel Javanmard, Karthik C. S., Ken-ichi Kawarabayashi, Ravi Kumar, Silvio Lattanzi, Euiwoong Lee, Yi Li, Ioannis Panageas, Dimitris Paparas, Benjamin Przybocki, Bernardo Subercaseaux, Ola Svensson, Shayan Taherijam, Xuan Wu, Eylon Yogev, Morteza Zadimoghaddam, Samson Zhou, Yossi Matias, James Manyika, Vahab Mirrokni

Computation and Language · Cross-listed from cs.AI

AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations

Minjun Zhu, Zhen Lin, Yixuan Weng et al. (+6 more)

High-quality scientific illustrations are crucial for effectively communicating complex scientific and technical concepts, yet their manual creation remains a well-recognized bottleneck in both academia and industry. We present FigureBench, the first large-scale benchmark for generating scientific illustrations from long-form scientific texts. It contains 3,300 high-quality scientific text-figure pairs, covering diverse text-to-illustration tasks from scientific papers, surveys, blogs, and textbooks. Moreover, we propose AutoFigure, the first agentic framework that automatically generates high-quality scientific illustrations based on long-form scientific text. Specifically, before rendering the final result, AutoFigure engages in extensive thinking, recombination, and validation to produce a layout that is both structurally sound and aesthetically refined, outputting a scientific illustration that achieves both structural completeness and aesthetic appeal. Leveraging the high-quality data from FigureBench, we conduct extensive experiments to test the performance of AutoFigure against various baseline methods. The results demonstrate that AutoFigure consistently surpasses all baseline methods, producing publication-ready scientific illustrations. The code, dataset and huggingface space are released in https://github.com/ResearAI/AutoFigure.
Read the full abstract

All authors Minjun Zhu, Zhen Lin, Yixuan Weng, Panzhong Lu, Qiujie Xie, Yifan Wei, Sifan Liu, Qiyao Sun, Yue Zhang

Computation and Language

They Said Memes Were Harmless-We Found the Ones That Hurt: Decoding Jokes, Symbols, and Cultural References

Sahil Tripathi, Gautam Siddharth Kashyap, Mehwish Nasim et al. (+3 more)

Meme-based social abuse detection is challenging because harmful intent often relies on implicit cultural symbolism and subtle cross-modal incongruence. Prior approaches, from fusion-based methods to in-context learning with Large Vision-Language Models (LVLMs), have made progress but remain limited by three factors: i) cultural blindness (missing symbolic context), ii) boundary ambiguity (satire vs. abuse confusion), and iii) lack of interpretability (opaque model reasoning). We introduce CROSS-ALIGN+, a three-stage framework that systematically addresses these limitations: (1) Stage I mitigates cultural blindness by enriching multimodal representations with structured knowledge from ConceptNet, Wikidata, and Hatebase; (2) Stage II reduces boundary ambiguity through parameter-efficient LoRA adapters that sharpen decision boundaries; and (3) Stage III enhances interpretability by generating cascaded explanations. Extensive experiments on five benchmarks and eight LVLMs demonstrate that CROSS-ALIGN+ consistently outperforms state-of-the-art methods, achieving up to 17% relative F1 improvement while providing interpretable justifications for each decision.
Read the full abstract

All authors Sahil Tripathi, Gautam Siddharth Kashyap, Mehwish Nasim, Jian Yang, Jiechao Gao, Usman Naseem

Computation and Language · Cross-listed from cs.LG

Antidistillation Fingerprinting

Yixuan Even Xu, John Kirchenbauer, Yash Savani et al. (+5 more)

Model distillation enables efficient emulation of frontier large language models (LLMs), creating a need for robust mechanisms to detect when a third-party student model has trained on a teacher model's outputs. However, existing fingerprinting techniques that could be used to detect such distillation rely on heuristic perturbations that impose a steep trade-off between generation quality and fingerprinting strength, often requiring significant degradation of utility to ensure the fingerprint is effectively internalized by the student. We introduce antidistillation fingerprinting (ADFP), a principled approach that aligns the fingerprinting objective with the student's learning dynamics. Building upon the gradient-based framework of antidistillation sampling, ADFP utilizes a proxy model to identify and sample tokens that directly maximize the expected detectability of the fingerprint in the student after fine-tuning, rather than relying on the incidental absorption of the un-targeted biases of a more naive watermark. Experiments on GSM8K, OASST1, and MBPP demonstrate that ADFP achieves a significant Pareto improvement over state-of-the-art baselines, yielding stronger detection confidence with minimal impact on utility across mathematical reasoning, dialogue, and code generation, even when the student model's architecture is unknown.
Read the full abstract

All authors Yixuan Even Xu, John Kirchenbauer, Yash Savani, Asher Trockman, Alexander Robey, Tom Goldstein, Fei Fang, J. Zico Kolter

Computation and Language · Cross-listed from cs.LG

COBALT: Contextual Bandit Learning with Offline Trajectories for Multi-Turn Code Generation

Ziru Chen, Dongdong Chen, Ruinan Jin et al. (+3 more)

Training large language models (LLMs) for multi-turn code generation typically relies on online reinforcement learning (RL), in which models learn from repeatedly generated online rollouts. Although effective, online RL can be costly and unstable. In this paper, we present contextual bandit learning with offline trajectories (COBALT), a simple method that reuses offline multi-turn trajectories to efficiently train LLMs for multi-turn code generation. Building on the observation that multi-turn code generation is a one-step recoverable Markov decision process, COBALT first collects code generation trajectories using a reference LLM and divides them into partial trajectories. Then, during online bandit learning, the LLM is trained to complete each partial trajectory through single-step code generation. On LiveCodeBench, COBALT substantially improves R1-Distill 8B and Qwen3 8B by up to 9.0 and 6.2 absolute Pass@1 points and matches the performance of two multi-turn online RL baselines using GRPO and VeRPO. We also perform a perturbation-based analysis of in-context reward hacking and show that Cobalt can improve robustness to misleading execution feedback. Overall, our results demonstrate that Cobalt is a promising solution for multi-turn code generation. Our code and data are available at https://github.com/OSU-NLP-Group/cobalt.
Read the full abstract

All authors Ziru Chen, Dongdong Chen, Ruinan Jin, Yingbin Liang, Yujia Xie, Huan Sun

Computation and Language · Cross-listed from cs.SE

FullStack-Agent: Enhancing Agentic Full-Stack Web Coding via Development-Oriented Testing and Repository Back-Translation

Zimu Lu, Houxing Ren, Yunqiao Yang et al. (+4 more)

Assisting non-expert users to develop complex interactive websites has become a popular task for LLM-powered code agents. However, existing code agents tend to only generate frontend web pages, masking the lack of real full-stack data processing and storage with fancy visual effects. Notably, constructing production-level full-stack web applications is far more challenging than only generating frontend web pages, demanding careful control of data flow, comprehensive understanding of constantly updating packages and dependencies, and accurate localization of obscure bugs in the codebase. To address these difficulties, we introduce FullStack-Agent, a unified agent system for full-stack agentic coding that consists of three parts: (1) FullStack-Dev, a multi-agent framework with strong planning, code editing, codebase navigation, and bug localization abilities. (2) FullStack-Learn, an innovative data-scaling and self-improving method that back-translates crawled and synthesized website repositories to improve the backbone LLM of FullStack-Dev. (3) FullStack-Bench, a comprehensive benchmark that systematically tests the frontend, backend and database functionalities of the generated website. Our FullStack-Dev outperforms the previous state-of-the-art method by 8.7%, 38.2%, and 15.9% on the frontend, backend, and database test cases respectively. Additionally, FullStack-Learn raises the performance of a 30B model by 9.7%, 9.5%, and 2.8% on the three sets of test cases through self-improvement, demonstrating the effectiveness of our approach. The code is released at https://github.com/mnluzimu/FullStack-Agent.
Read the full abstract

All authors Zimu Lu, Houxing Ren, Yunqiao Yang, Ke Wang, Zhuofan Zong, Mingjie Zhan, Hongsheng Li

Computation and Language · Cross-listed from cs.CR

WebSentinel: Detecting and Localizing Prompt Injection Attacks for Web Agents

Xilong Wang, Yinuo Liu, Zhun Wang et al. (+2 more)

Prompt injection attacks manipulate webpage content to cause web agents to execute attacker-specified tasks instead of the user's intended ones. Existing methods for detecting and localizing such attacks achieve limited effectiveness, as their underlying assumptions often do not hold in the web-agent setting. In this work, we propose WebSentinel, a two-step approach for detecting and localizing prompt injection attacks in webpages. Given a webpage, Step I extracts segments of interest that may be contaminated, and Step II evaluates each segment by checking its consistency with the webpage content as context. We show that WebSentinel is highly effective, substantially outperforming baseline methods across multiple datasets of both contaminated and clean webpages that we collected. Our code is available at: https://github.com/wxl-lxw/WebSentinel.
Read the full abstract

All authors Xilong Wang, Yinuo Liu, Zhun Wang, Dawn Song, Neil Gong

Computation and Language · Cross-listed from cs.AI

AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration

Jianhao Ruan, Zhihao Xu, Yiran Peng et al. (+9 more)

Language agents have shown strong promise for task automation. Realizing this promise for increasingly complex, long-horizon tasks has driven the rise of a sub-agent-as-tools paradigm for multi-turn task solving. However, existing designs still lack a dynamic abstraction view of sub-agents, thereby hurting adaptability. We address this challenge with a unified, framework-agnostic agent abstraction that models any agent as a tuple Instruction, Context, Tools, Model. This tuple acts as a compositional recipe for capabilities, enabling the system to spawn specialized executors for each task on demand. Building on this abstraction, we introduce an agentic system AOrchestra, where the central orchestrator concretizes the tuple at each step: it curates task-relevant context, selects tools and models, and delegates execution via on-the-fly automatic agent creation. Such designs enable reducing human engineering efforts, and remain framework-agnostic with plug-and-play support for diverse agents as task executors. It also enables a controllable performance-cost trade-off, allowing the system to approach Pareto-efficient. Across three challenging benchmarks (GAIA, SWE-Bench, Terminal-Bench), AOrchestra achieves 16.28% relative improvement against the strongest baseline when paired with Gemini-3-Flash. The code is available at: https://github.com/FoundationAgents/AOrchestra
Read the full abstract

All authors Jianhao Ruan, Zhihao Xu, Yiran Peng, Fashen Ren, Zhaoyang Yu, Xinbing Liang, Jinyu Xiang, Yongru Chen, Bang Liu, Chenglin Wu, Yuyu Luo, Jiayi Zhang

Computation and Language

Fix the Structural Bottleneck: Context Compression via Explicit Information Transmission

Jiangnan Ye, Hanqi Yan, Zhenyi Shen et al. (+3 more)

Long-context LLM agents often struggle with growing token, memory, and latency costs, making efficient context compression essential for practical deployment. Existing LLM-as-a-compressor methods remain noticeably inferior to using the full context. We find that this gap partly stems from their inability to preserve contextual information effectively. In this work, we revisit context compression from a structural perspective and identify two key bottlenecks in standard LLM-based compressors: limited coordination among compression tokens during information aggregation, and layerwise dilution that weakens useful signals from intermediate hidden states. To address these limitations, we propose ComprExIT, a new context compression framework based on explicit information transmission. ComprExIT adaptively selects features across frozen LLM layers, then allocates information from anchors to compression slots through a globally coordinated transport plan. Experiments on 12 datasets show that ComprExIT consistently outperforms strong soft-compression baselines, improving average F1 by up to 18.5%, while adding only ~1% trainable parameters and achieving more than 2x faster compression than the fastest baselines. The code will be released upon acceptance.
Read the full abstract

All authors Jiangnan Ye, Hanqi Yan, Zhenyi Shen, Heng Chang, Ye Mao, Yulan He

Computation and Language · Cross-listed from cs.LG

Efficient Estimation of Kernel Surrogate Models for Task Attribution

Zhenshuo Zhang, Minxuan Duan and Hongyang R. Zhang

Modern AI agents such as large language models are trained on diverse tasks – translation, code generation, mathematical reasoning, and text prediction – simultaneously. A key question is how to quantify the influence of each individual training task on performance on a target task, a problem we refer to as task attribution. The direct approach, leave-one-out retraining, measures the effect of removing each task, but is computationally infeasible at scale. An alternative approach that builds surrogate models to predict the performance on a target task for any subset of training tasks has emerged in the recent literature. Prior work focuses on linear surrogate models, which capture first-order relationships but miss nonlinear interactions such as XOR-type effects. In this paper, we first consider a unified task-weighting framework for analyzing task-attribution methods and establish a new connection between linear surrogate models and influence functions via a second-order analysis. Then, we introduce kernel surrogate models, which more effectively represent second-order task interactions. To efficiently learn the kernel surrogate, we develop a gradient-based estimation procedure that leverages a first-order approximation of pretrained models; empirically, this yields accurate surrogate estimates with less than $2\%$ relative error without repeated retraining. Experiments across multiple settings – including mathematical reasoning in transformers, in-context learning, and multi-objective reinforcement learning – demonstrate the effectiveness of kernel surrogate models. They achieve a $25\%$ higher correlation with the leave-one-out ground truth than linear surrogates and influence-function baselines, enabling more accurate and scalable task attribution. When used for downstream data selection, kernel surrogate models further yield a $40\%$ improvement in the aforementioned settings.
Read the full abstract

Computation and Language

CUBO: Self-Contained Retrieval-Augmented Generation on Consumer Laptops 10 GB Corpora, 16 GB RAM, Single-Device Deployment

Paolo Astrino

Organizations handling sensitive documents face a tension: cloud-based AI risks GDPR violations, while local systems typically require 18-32 GB RAM. This paper presents CUBO, a systems-oriented RAG platform for consumer laptops with 16 GB shared memory. CUBO's novelty lies in engineering integration of streaming ingestion (O(1) buffer overhead), tiered hybrid retrieval, and hardware-aware orchestration that enables competitive Recall@10 (0.48-0.97 across BEIR domains) within a hard 15.5 GB RAM ceiling. The 37,000-line codebase achieves retrieval latencies of 185 ms (p50) on C1,300 laptops while maintaining data minimization through local-only processing aligned with GDPR Art. 5(1)(c). Evaluation on BEIR benchmarks validates practical deployability for small-to-medium professional archives. The codebase is publicly available at https://github.com/PaoloAstrino/CUBO.
Read the full abstract

Computation and Language

BranPO: Scalable Contrastive Branch Sampling for Long-Horizon Agentic Reinforcement Learning

Yubao Zhao, Weiquan Huang, Sudong Wang et al. (+4 more)

Agentic reinforcement learning enables large language models to perform multi-turn planning and tool use, but long-horizon training remains challenging under sparse trajectory-level rewards, where a single outcome is uniformly assigned to all decisions. Prior methods introduce finer-grained supervision via tree-based exploration or process-level evaluation, but often incur high cost or produce noisy credit signals. In agentic trajectories, early mistakes may still be corrected by later actions, while seemingly promising intermediate states can fail due to poor subsequent decisions. We call this property non-monotonic correctness, which makes outcome rewards or state values insufficient for guiding what actions should be taken from each state. To address this, we propose Branching Relative Policy Optimization (BranPO), a value-free method that constructs localized contrastive supervision without dense rewards. BranPO truncates trajectories at intermediate prefixes and resamples continuations to form contrastive branches that share the same prefix but diverge in final outcomes, thereby isolating decisions that drive success or failure. We further introduce difficulty-aware branch sampling and Redundant Step Masking to improve sampling efficiency and suppress redundant updates. Experiments show that BranPO consistently outperforms diverse baseline categories across multiple multi-hop QA benchmarks without additional training cost, and generalizes to broader long-horizon agentic tasks with consistent improvements. Our code is available at https://github.com/YubaoZhao/BranPO.
Read the full abstract

All authors Yubao Zhao, Weiquan Huang, Sudong Wang, Ruochen Zhao, Chen Chen, Yao Shu, Chengwei Qin

Computation and Language

No Shortcuts to Culture: Indonesian Multi-hop Question Answering for Complex Cultural Understanding

Vynska Amalia Permadi, Xingwei Tan, Nafise Sadat Moosavi et al. (+1 more)

Understanding culture requires reasoning across context, tradition, and implicit social knowledge, far beyond recalling isolated facts. Yet most culturally focused question answering (QA) benchmarks rely on single-hop questions, which may allow models to exploit shallow cues rather than demonstrate genuine cultural reasoning. In this work, we introduce ID-MoCQA, the first large-scale multi-hop QA dataset for assessing the cultural understanding of large language models (LLMs), grounded in Indonesian traditions and available in both English and Indonesian. We present a new framework that systematically transforms single-hop cultural questions into multi-hop reasoning chains spanning six clue types (e.g., commonsense, temporal, geographical). Our multi-stage validation pipeline, combining expert review and LLM-as-a-judge filtering, ensures high-quality question-answer pairs. Our evaluation across state-of-the-art models reveals substantial gaps in cultural reasoning, particularly in tasks requiring nuanced inference. ID-MoCQA provides a challenging and essential benchmark for advancing the cultural competency of LLMs.
Read the full abstract

All authors Vynska Amalia Permadi, Xingwei Tan, Nafise Sadat Moosavi, Nikos Aletras

Computation and Language

Beyond Tokens: Semantic-Aware Speculative Decoding for Efficient Inference by Probing Internal States

Ximing Dong, Shaowei Wang, Dayi Lin et al. (+2 more)

Large Language Models (LLMs) achieve strong performance across many tasks but suffer from high inference latency due to autoregressive decoding. The issue is exacerbated in Large Reasoning Models (LRMs), which generate lengthy chains of thought. While speculative decoding accelerates inference by drafting and verifying multiple tokens in parallel, existing methods operate at the token level and ignore semantic equivalence (i.e., different token sequences expressing the same meaning), leading to inefficient rejections. We propose SemanticSpec, a semantic-aware speculative decoding framework that verifies entire semantic sequences instead of tokens. SemanticSpec introduces a semantic probability estimation mechanism that probes the model's internal hidden states to assess the likelihood of generating sequences with specific meanings. Experiments on four benchmarks show that SemanticSpec achieves up to 2.7x speedup on DeepSeekR1-32B and 2.1x on QwQ-32B, consistently outperforming token-level and sequence-level baselines in both efficiency and effectiveness.
Read the full abstract

All authors Ximing Dong, Shaowei Wang, Dayi Lin, Boyuan Chen, Ahmed E. Hassan

Computation and Language

OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering

Yifan Zhu, Xinyu Mu, Tao Feng et al. (+3 more)

Long-horizon omnimodal question answering answers questions by reasoning over text, images, audio, and video. Despite recent progress on OmniLLMs, low-resource long audio-video QA still suffers from costly dense encoding, weak fine-grained retrieval, limited proactive planning, and no clear end-to-end optimization. To address these issues, we propose OmniRAG-Agent, an agentic omnimodal QA method for budgeted long audio-video reasoning. It builds an image-audio retrieval-augmented generation module that lets an OmniLLM fetch short, relevant frames and audio snippets from external banks. Moreover, it uses an agent loop that plans, calls tools across turns, and merges retrieved evidence to answer complex queries. Furthermore, we apply group relative policy optimization to jointly improve tool use and answer quality over time. Experiments on OmniVideoBench, WorldSense, and Daily-Omni show that OmniRAG-Agent consistently outperforms prior methods under low-resource settings and achieves strong results, with ablations validating each component.
Read the full abstract

All authors Yifan Zhu, Xinyu Mu, Tao Feng, Zhonghong Ou, Yuning Gong, Haoran Luo

Computation and Language

Cognitively Diverse Multiple-Choice Question Generation: A Hybrid Multi-Agent Framework with Large Language Models

Yu Tian, Linh Huynh, Katerina Christhilf et al. (+4 more)

Recent advances in large language models (LLMs) have made automated multiple-choice question (MCQ) generation increasingly feasible; however, reliably producing items that satisfy controlled cognitive demands remains a challenge. To address this gap, we introduce ReQUESTA, a hybrid, multi-agent framework for generating cognitively diverse MCQs that systematically target text-based, inferential, and main idea comprehension. ReQUESTA decomposes MCQ authoring into specialized subtasks and coordinates LLM-powered agents with rule-based components to support planning, controlled generation, iterative evaluation, and post-processing. We evaluated the framework in a large-scale reading comprehension study using academic expository texts, comparing ReQUESTA-generated MCQs with those produced by a single-pass GPT-5 zero-shot baseline. Psychometric analyses of learner responses assessed item difficulty and discrimination, while expert raters evaluated question quality across multiple dimensions, including topic relevance and distractor quality. Results showed that ReQUESTA-generated items were consistently more challenging, more discriminative, and more strongly aligned with overall reading comprehension performance. Expert evaluations further indicated stronger alignment with central concepts and superior distractor linguistic consistency and semantic plausibility, particularly for inferential questions. These findings demonstrate that hybrid, agentic orchestration can systematically improve the reliability and controllability of LLM-based generation, highlighting workflow design as a key lever for structured artifact generation beyond single-pass prompting.
Read the full abstract

All authors Yu Tian, Linh Huynh, Katerina Christhilf, Shubham Chakraborty, Micah Watanabe, Tracy Arner, Danielle McNamara

Computation and Language · Cross-listed from cs.LG

Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updates

Duy Nguyen, Hanqi Xiao, Archiki Prasad et al. (+3 more)

Large language models (LLMs) rely on internal knowledge to solve many downstream tasks, making it crucial to keep them up to date. Since full retraining is expensive, prior work has explored efficient alternatives such as model editing and parameter-efficient fine-tuning. However, these approaches often break down in practice due to poor generalization across inputs, limited stability, and knowledge conflict. To address these limitations, we propose the CoRSA (Conflict-Resolving and Sharpness-Aware Minimization) training framework, a parameter-efficient, holistic approach for knowledge editing with multiple updates. CoRSA tackles multiple challenges simultaneously: it improves generalization to different input forms and enhances stability across multiple updates by minimizing loss curvature, and resolves conflicts by maximizing the margin between new and prior knowledge. Across three widely used fact editing benchmarks, CoRSA achieves significant gains in generalization, outperforming baselines with average absolute improvements of 12.42% over LoRA and 10% over model editing methods. With multiple updates, it maintains high update efficacy while reducing catastrophic forgetting by 27.82% compared to LoRA. CoRSA also generalizes to the code domain, outperforming the strongest baseline by 5.48% Pass@5 in update efficacy.
Read the full abstract

All authors Duy Nguyen, Hanqi Xiao, Archiki Prasad, Elias Stengel-Eskin, Hyunji Lee, Mohit Bansal

Computation and Language · Cross-listed from cs.MA

Agent Primitives: Reusable Latent Building Blocks for Multi-Agent Systems

Haibo Jin, Peng Kuang, Ye Yu et al. (+2 more)

While existing multi-agent systems (MAS) can handle complex problems by enabling collaboration among multiple agents, they are often highly task-specific, relying on manually crafted agent roles and interaction prompts, which leads to increased architectural complexity and limited reusability across tasks. Moreover, most MAS communicate primarily through natural language, making them vulnerable to error accumulation and instability in long-context, multi-stage interactions within internal agent histories. In this work, we propose Agent Primitives, a set of reusable latent building blocks for LLM-based MAS. Inspired by neural network design, where complex models are built from reusable components, we observe that many existing MAS architectures can be decomposed into a small number of recurring internal computation patterns. Based on this observation, we instantiate three primitives: Review, Voting and Selection, and Planning and Execution. All primitives communicate internally via key-value (KV) cache, which improves both robustness and efficiency by mitigating information degradation across multi-stage interactions. To enable automatic system construction, an Organizer agent selects and composes primitives for each query, guided by a lightweight knowledge pool of previously successful configurations, forming a primitive-based MAS. Experiments show that primitives-based MAS improve average accuracy by 12.0-16.5% over single-agent baselines, reduce token usage and inference latency by approximately 3$\times$-4$\times$ compared to text-based MAS, while incurring only 1.3$\times$-1.6$\times$ overhead relative to single-agent inference and providing more stable performance across model backbones.
Read the full abstract

All authors Haibo Jin, Peng Kuang, Ye Yu, Xiaopeng Yuan, Haohan Wang

Computation and Language

OCRTurk: A Comprehensive OCR Benchmark for Turkish

Deniz Y{\i}lmaz, Evren Ayberk Munis, Çağr{\i} Toraman et al. (+4 more)

Document parsing is now widely used in applications, such as large-scale document digitization, retrieval-augmented generation, and domain-specific pipelines in healthcare and education. Benchmarking these models is crucial for assessing their reliability and practical robustness. Existing benchmarks mostly target high-resource languages and provide limited coverage for low-resource settings, such as Turkish. Moreover, existing studies on Turkish document parsing lack a standardized benchmark that reflects real-world scenarios and document diversity. To address this gap, we introduce OCRTurk, a Turkish document parsing benchmark covering multiple layout elements and document categories at three difficulty levels. OCRTurk consists of 180 Turkish documents drawn from academic articles, theses, slide decks, and non-academic articles. We evaluate seven OCR models on OCRTurk using element-wise metrics. Across difficulty levels, PaddleOCR achieves the strongest overall results, leading most element-wise metrics except figures and attaining high Normalized Edit Distance scores in easy, medium, and hard subsets. We also observe performance variation by document type. Models perform well on non-academic documents, while slideshows become the most challenging.
Read the full abstract

All authors Deniz Y{\i}lmaz, Evren Ayberk Munis, Çağr{\i} Toraman, Süha Kağan Köse, Burak Aktaş, Mehmet Can Baytekin, Bilge Kaan Görür

Computation and Language

Rethinking the Reranker: Boundary-Aware Evidence Selection for Robust Retrieval-Augmented Generation

Jiashuo Sun, Pengcheng Jiang, Saizhuo Wang et al. (+13 more)

Retrieval-Augmented Generation (RAG) systems remain brittle under realistic retrieval noise, even when the required evidence appears in the top-K results. A key reason is that retrievers and rerankers optimize solely for relevance, often selecting either trivial, answer-revealing passages or evidence that lacks the critical information required to answer the question, without considering whether the evidence is suitable for the generator. We propose BAR-RAG, which reframes the reranker as a boundary-aware evidence selector that targets the generator's Goldilocks Zone – evidence that is neither trivially easy nor fundamentally unanswerable for the generator, but is challenging yet sufficient for inference and thus provides the strongest learning signal. BAR-RAG trains the selector with reinforcement learning using generator feedback, and adopts a two-stage pipeline that fine-tunes the generator under the induced evidence distribution to mitigate the distribution mismatch between training and inference. Experiments on knowledge-intensive question answering benchmarks show that BAR-RAG consistently improves end-to-end performance under noisy retrieval, achieving an average gain of 10.3 percent over strong RAG and reranking baselines while substantially improving robustness. Code is publicly avaliable at https://github.com/GasolSun36/BAR-RAG.
Read the full abstract

All authors Jiashuo Sun, Pengcheng Jiang, Saizhuo Wang, Jiajun Fan, Heng Wang, Siru Ouyang, Ming Zhong, Yizhu Jiao, Chengsong Huang, Xueqiang Xu, Pengrui Han, Peiran Li, Jiaxin Huang, Ge Liu, Heng Ji, Jiawei Han

Computation and Language

Neural Attention Search Linear: Towards Adaptive Token-Level Hybrid Attention Models

Difan Deng, Andreas Bentzen Winje, Lukas Fehring et al. (+1 more)

The quadratic computational complexity of softmax transformers has become a bottleneck in long-context scenarios. In contrast, linear attention model families provide a promising direction towards a more efficient sequential model. These linear attention models compress past KV values into a single hidden state, thereby efficiently reducing complexity during both training and inference. However, their expressivity remains limited by the size of their hidden state. Previous work proposed interleaving softmax and linear attention layers to reduce computational complexity while preserving expressivity. Nevertheless, the efficiency of these models remains bottlenecked by their softmax attention layers. In this paper, we propose Neural Attention Search Linear (NAtS-L), a framework that applies both linear attention and softmax attention operations within the same layer on different tokens. NAtS-L automatically determines whether a token can be handled by a linear attention model, i.e., tokens that have only short-term impact and can be encoded into fixed-size hidden states, or require softmax attention, i.e., tokens that contain information related to long-term retrieval and need to be preserved for future queries. By searching for optimal Gated DeltaNet and softmax attention combinations across tokens, we show that NAtS-L provides a strong yet efficient token-level hybrid architecture.
Read the full abstract

All authors Difan Deng, Andreas Bentzen Winje, Lukas Fehring, Marius Lindauer

Computation and Language

Instruction Anchor: Dissecting the Mechanistic Dynamics of Modality Arbitration

Yu Zhang, Mufan Xu, Xuefeng Bai et al. (+4 more)

Modality following is the ability to selectively leverage multimodal contexts based on user instructions. It is fundamental to the safety and reliability of multimodal large language models (MLLMs) in real-world deployments. However, the internal mechanisms governing this decision-making process remain largely under-explored. In this work, we investigate the mechanism underlying modality following through an information flow perspective. Our findings reveal that instruction tokens serve as structural anchor for modality arbitration: Shallow attention layers perform undifferentiated information transfer, aggregating multimodal cues to instruction tokens as a latent buffer; in contrast, deep attention layers selectively strengthen the instruction-compliant subspace and resolve modality arbitration according to the instruction-specified intent, with a sparse subset of attention heads driving this process. Targeted attention-head interventions further validate the functional specificity of these heads: blocking only $5\%$ of the identified heads substantially degrades modality following while preserving general visual and language capabilities, whereas targeted amplification can restore failed modality-following samples by up to approximately $60\%$. Together, this work provides a mechanistic account of modality following and informs future efforts to improve how MLLMs integrate and utilize multimodal evidence under user instructions.
Read the full abstract

All authors Yu Zhang, Mufan Xu, Xuefeng Bai, Kehai Chen, Pengfei Zhang, Yang Xiang, Min Zhang