Peer Revenue Equity in science

Papers

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

Computer Science Computation and Language

cs.CL Watch Computation and Language

Wednesday 18 March

Computation and Language

How Psychological Learning Paradigms Shaped and Constrained Artificial Intelligence

Alex Anvi Eponon, Ildar Batyrshin, Christian E. Maldonado-Sifuentes et al. (+1 more)

Current artificial intelligence systems struggle with systematic compositional reasoning: the capacity to recombine known components in novel configurations. This paper argues that the failure is architectural, not merely a matter of scale or training data, and that its origins lie in the psychological learning theories from which AI paradigms were derived. The argument proceeds in three stages. First, drawing on the systematicity debate in cognitive science and on the demonstration of Aizawa that neither connectionism nor classicism can make systematicity a structural consequence of the architecture, the paper establishes that the corrective techniques proliferating in modern AI, from chain-of-thought prompting to alignment through human feedback, function as auxiliary hypotheses that address symptoms without resolving the underlying architectural indifference to systematicity. Second, it traces the genealogy from psychological learning theory to AI methodology, showing that behaviourism, cognitivism, and constructivism each bequeathed a specific structural limitation to the AI paradigm it inspired: the exclusion of internal structure, the opacity of representation, and the absence of formal construction operators. A cross-cultural reappraisal of rote learning reveals a further underexploited pathway. Third, the paper introduces ReSynth, a trimodular conceptual framework that proposes the principled separation of reasoning, identity, and memory as a path toward architectures in which systematic behaviour is a structural consequence of design rather than a correction applied after the fact.
Read the full abstract

All authors Alex Anvi Eponon, Ildar Batyrshin, Christian E. Maldonado-Sifuentes, Grigori Sidorov

Computation and Language

CWoMP: Morpheme Representation Learning for Interlinear Glossing

Morris Alper, Enora Rice, Bhargav Shandilya et al. (+2 more)

Interlinear glossed text (IGT) is a standard notation for language documentation which is linguistically rich but laborious to produce manually. Recent automated IGT methods treat glosses as character sequences, neglecting their compositional structure. We propose CWoMP (Contrastive Word-Morpheme Pretraining), which instead treats morphemes as atomic form-meaning units with learned representations. A contrastively trained encoder aligns words-in-context with their constituent morphemes in a shared embedding space; an autoregressive decoder then generates the morpheme sequence by retrieving entries from a mutable lexicon of these embeddings. Predictions are interpretable–grounded in lexicon entries–and users can improve results at inference time by expanding the lexicon without retraining. We evaluate on diverse low-resource languages, showing that CWoMP outperforms existing methods while being significantly more efficient, with particularly strong gains in extremely low-resource settings.
Read the full abstract

All authors Morris Alper, Enora Rice, Bhargav Shandilya, Alexis Palmer, Lori Levin

Computation and Language

GRAFITE: Generative Regression Analysis Framework for Issue Tracking and Evaluation

Ja Young Lee, Mírian Silva, Mohamed Nasr et al. (+6 more)

Large language models (LLMs) are largely motivated by their performance on popular topics and benchmarks at the time of their release. However, over time, contamination occurs due to significant exposure of benchmark data during training. This poses a risk of model performance inflation if testing is not carefully executed. To address this challenge, we present GRAFITE, a continuous LLM evaluation platform through a comprehensive system for maintaining and evaluating model issues. Our approach enables building a repository of model problems based on user feedback over time and offers a pipeline for assessing LLMs against these issues through quality assurance (QA) tests using LLM-as-a-judge. The platform enables side-by-side comparison of multiple models, facilitating regression detection across different releases. The platform is available at https://github.com/IBM/grafite. The demo video is available at www.youtube.com/watch?v=XFZyoleN56k.
Read the full abstract

All authors Ja Young Lee, Mírian Silva, Mohamed Nasr, Shonda Witherspoon, Enzo Bozzani, Veronique Demers, Radha Ratnaparkhi, Hui Wu, Sara Rosenthal

Computation and Language

Modeling the human lexicon under temperature variations: linguistic factors, diversity and typicality in LLM word associations

Maria Andueza Rodriguez, Marie Candito and Richard Huyghe

Large language models (LLMs) achieve impressive results in terms of fluency in text generation, yet the nature of their linguistic knowledge - in particular the human-likeness of their internal lexicon - remains uncertain. This study compares human and LLM-generated word associations to evaluate how accurately models capture human lexical patterns. Using English cue-response pairs from the SWOW dataset and newly generated associations from three LLMs (Mistral-7B, Llama-3.1-8B, and Qwen-2.5-32B) across multiple temperature settings, we examine (i) the influence of lexical factors such as word frequency and concreteness on cue-response pairs, and (ii) the variability and typicality of LLM responses compared to human responses. Results show that all models mirror human trends for frequency and concreteness but differ in response variability and typicality. Larger models such as Qwen tend to emulate a single "prototypical" human participant, generating highly typical but minimally variable responses, while smaller models such as Mistral and Llama produce more variable yet less typical responses. Temperature settings further influence this trade-off, with higher values increasing variability but decreasing typicality. These findings highlight both the similarities and differences between human and LLM lexicons, emphasizing the need to account for model size and temperature when probing LLM lexical representations.
Read the full abstract

Computation and Language

How LLMs Distort Our Written Language

Marwa Abdulhai, Isadora White, Yanming Wan et al. (+4 more)

Large language models (LLMs) are used by over a billion people globally, most often to assist with writing. In this work, we demonstrate that LLMs not only alter the voice and tone of human writing but also consistently alter the intended meaning. First, we conduct a human user study to understand how people actually interact with LLMs when using them for writing. Our findings reveal that extensive LLM use led to a nearly 70% increase in essays that remained neutral in answering the topic question. Significantly more heavy LLM users reported that the writing was less creative and not in their voice. Next, using a dataset of human-written essays that was collected in 2021 before the widespread release of LLMs, we study how asking an LLM to revise the essay based on the human-written feedback in the dataset induces large changes in the resulting content and meaning. We find that even when LLMs are prompted with expert feedback and asked to only make grammar edits, they still change the text in a way that significantly alters its semantic meaning. We then examine LLM-generated text in the wild, specifically focusing on the 21% of AI-generated scientific peer reviews at a recent top AI conference. We find that LLM-generated reviews place significantly less weight on clarity and significance of the research, and assign scores that, on average, are a full point higher. These findings highlight a misalignment between the perceived benefit of AI use and an implicit, consistent effect on the semantics of human writing, motivating future work on how widespread AI writing will affect our cultural and scientific institutions.
Read the full abstract

All authors Marwa Abdulhai, Isadora White, Yanming Wan, Ibrahim Qureshi, Joel Z. Leibo, Max Kleiman-Weiner, Natasha Jaques

Computation and Language

Evaluating FrameNet-Based Semantic Modeling for Gender-Based Violence Detection in Clinical Records

Lívia Dutra, Arthur Lorenzi, Frederico Belcavello et al. (+8 more)

Gender-based violence (GBV) is a major public health issue, with the World Health Organization estimating that one in three women experiences physical or sexual violence by an intimate partner during her lifetime. In Brazil, although healthcare professionals are legally required to report such cases, underreporting remains significant due to difficulties in identifying abuse and limited integration between public information systems. This study investigates whether FrameNet-based semantic annotation of open-text fields in electronic medical records can support the identification of patterns of GBV. We compare the performance of an SVM classifier for GBV cases trained on (1) frame-annotated text, (2) annotated text combined with parameterized data, and (3) parameterized data alone. Quantitative and qualitative analyses show that models incorporating semantic annotation outperform categorical models, achieving over 0.3 improvement in F1 score and demonstrating that domain-specific semantic representations provide meaningful signals beyond structured demographic data. The findings support the hypothesis that semantic analysis of clinical narratives can enhance early identification strategies and support more informed public health interventions.
Read the full abstract

All authors Lívia Dutra, Arthur Lorenzi, Frederico Belcavello, Ely Matos, Marcelo Viridiano, Lorena Larré, Olívia Guaranha, Erik Santos, Sofia Reinach, Pedro de Paula, Tiago Torrent

Computation and Language · Cross-listed from cs.SD

MOSS-TTS Technical Report

Yitian Gong, Botian Jiang, Yiwei Zhao et al. (+23 more)

This technical report presents MOSS-TTS, a speech generation foundation model built on a scalable recipe: discrete audio tokens, autoregressive modeling, and large-scale pretraining. Built on MOSS-Audio-Tokenizer, a causal Transformer tokenizer that compresses 24 kHz audio to 12.5 fps with variable-bitrate RVQ and unified semantic-acoustic representations, we release two complementary generators: MOSS-TTS, which emphasizes structural simplicity, scalability, and long-context/control-oriented deployment, and MOSS-TTS-Local-Transformer, which introduces a frame-local autoregressive module for higher modeling efficiency, stronger speaker preservation, and a shorter time to first audio. Across multilingual and open-domain settings, MOSS-TTS supports zero-shot voice cloning, token-level duration control, phoneme-/pinyin-level pronunciation control, smooth code-switching, and stable long-form generation. This report summarizes the design, training recipe, and empirical characteristics of the released models.
Read the full abstract

All authors Yitian Gong, Botian Jiang, Yiwei Zhao, Yucheng Yuan, Kuangwei Chen, Yaozhou Jiang, Cheng Chang, Dong Hong, Mingshu Chen, Ruixiao Li, Yiyang Zhang, Yang Gao, Hanfu Chen, Ke Chen, Songlin Wang, Xiaogui Yang, Yuqian Zhang, Kexin Huang, ZhengYuan Lin, Kang Yu, Ziqi Chen, Jin Wang, Zhaoye Fei, Qinyuan Cheng, Shimin Li, Xipeng Qiu

Computation and Language · Cross-listed from cs.CV

Loc3R-VLM: Language-based Localization and 3D Reasoning with Vision-Language Models

Kevin Qu, Haozhe Qi, Mihai Dusmanu et al. (+3 more)

Multimodal Large Language Models (MLLMs) have made impressive progress in connecting vision and language, but they still struggle with spatial understanding and viewpoint-aware reasoning. Recent efforts aim to augment the input representations with geometric cues rather than explicitly teaching models to reason in 3D space. We introduce Loc3R-VLM, a framework that equips 2D Vision-Language Models with advanced 3D understanding capabilities from monocular video input. Inspired by human spatial cognition, Loc3R-VLM relies on two joint objectives: global layout reconstruction to build a holistic representation of the scene structure, and explicit situation modeling to anchor egocentric perspective. These objectives provide direct spatial supervision that grounds both perception and language in a 3D context. To ensure geometric consistency and metric-scale alignment, we leverage lightweight camera pose priors extracted from a pre-trained 3D foundation model. Loc3R-VLM achieves state-of-the-art performance in language-based localization and outperforms existing 2D- and video-based approaches on situated and general 3D question-answering benchmarks, demonstrating that our spatial supervision framework enables strong 3D understanding. Project page: https://kevinqu7.github.io/loc3r-vlm
Read the full abstract

All authors Kevin Qu, Haozhe Qi, Mihai Dusmanu, Mahdi Rad, Rui Wang, Marc Pollefeys

Computation and Language

ConGA: Guidelines for Contextual Gender Annotation. A Framework for Annotating Gender in Machine Translation

Argentina Anna Rescigno, Eva Vanmassenhove and Johanna Monti

Handling gender across languages remains a persistent challenge for Machine Translation (MT) and Large Language Models (LLMs), especially when translating from gender-neutral languages into morphologically gendered ones, such as English to Italian. English largely omits grammatical gender, while Italian requires explicit agreement across multiple grammatical categories. This asymmetry often leads MT systems to default to masculine forms, reinforcing bias and reducing translation accuracy. To address this issue, we present the Contextual Gender Annotation (ConGA) framework, a linguistically grounded set of guidelines for word-level gender annotation. The scheme distinguishes between semantic gender in English through three tags, Masculine (M), Feminine (F), and Ambiguous (A), and grammatical gender realisation in Italian (Masculine (M), Feminine (F)), combined with entity-level identifiers for cross-sentence tracking. We apply ConGA to the gENder-IT dataset, creating a gold-standard resource for evaluating gender bias in translation. Our results reveal systematic masculine overuse and inconsistent feminine realisation, highlighting persistent limitations of current MT systems. By combining fine-grained linguistic annotation with quantitative evaluation, this work offers both a methodology and a benchmark for building more gender-aware and multilingual NLP systems.
Read the full abstract

Computation and Language

Gender Disambiguation in Machine Translation: Diagnostic Evaluation in Decoder-Only Architectures

Chiara Manna, Hosein Mohebbi, Afra Alishahi et al. (+2 more)

While Large Language Models achieve state-of-the-art results across a wide range of NLP tasks, they remain prone to systematic biases. Among these, gender bias is particularly salient in MT, due to systematic differences across languages in whether and how gender is marked. As a result, translation often requires disambiguating implicit source signals into explicit gender-marked forms. In this context, standard benchmarks may capture broad disparities but fail to reflect the full complexity of gender bias in modern MT. In this paper, we extend recent frameworks on bias evaluation by: (i) introducing a novel measure coined "Prior Bias", capturing a model's default gender assumptions, and (ii) applying the framework to decoder-only MT models. Our results show that, despite their scale and state-of-the-art status, decoder-only models do not generally outperform encoder-decoder architectures on gender-specific metrics; however, post-training (e.g., instruction tuning) not only improves contextual awareness but also reduces the masculine Prior Bias.
Read the full abstract

All authors Chiara Manna, Hosein Mohebbi, Afra Alishahi, Frédéric Blain, Eva Vanmassenhove

Computation and Language

ShapleyLaw: A Game-Theoretic Approach to Multilingual Scaling Laws

Xuyang Cao, Qianying Liu, Chuan Xiao et al. (+7 more)

In multilingual pretraining, the test loss of a pretrained model is heavily influenced by the proportion of each language in the pretraining data, namely the language mixture ratios. Multilingual scaling laws can predict the test loss under different language mixture ratios and can therefore be used to estimate the optimal ratios. However, the current approaches to multilingual scaling laws do not measure the cross-lingual transfer effect, resulting in suboptimal mixture ratios. In this paper, we consider multilingual pretraining as a cooperative game in which each language acts as a player that jointly contributes to pretraining, gaining the resulting reduction in test loss as the payoff. Consequently, from the perspective of cooperative game theory, we quantify the cross-lingual transfer from each language by its contribution in the game, and propose a game-theoretic multilingual scaling law called ShapleyLaw. Our experiments show that ShapleyLaw outperforms baseline methods in model performance prediction and language mixture optimization.
Read the full abstract

All authors Xuyang Cao, Qianying Liu, Chuan Xiao, Yusuke Oda, Jiayi Wang, Pontus Stenetorp, Daisuke Kawahara, Makoto Onizuka, Sadao Kurohashi, Shuyuan Zheng

Computation and Language

Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing

Raghavv Goel, Mukul Gagrani, Mingu Lee et al. (+1 more)

Large Language Models (LLMs) possess latent multi-token prediction (MTP) abilities despite being trained only for next-token generation. We introduce ESP (Embedding-Space Probing), a simple and training-free MTP method that probes an LLM using on-the-fly mask tokens drawn from its embedding space, enabling parallel future-token prediction without modifying weights or relying on draft models. ESP constructs a speculative token tree by sampling Top-K candidates from mask-token logits and applies a lightweight pruning rule to retain high-probability continuations. During generation, predictions are verified in parallel, yielding lossless decoding while significantly reducing model calls and increasing token throughput. ESP consistently outperforms existing training-free baselines, improving acceptance length by 7-11% over LADE on LLaMA3 and 7-8% on Qwen3, and increasing throughput by up to 15-19% over the strongest baseline. Finally, we provide theoretical insight and empirical evidence showing that decoder layers naturally align mask-token representations with next-token states, enabling accurate multi-step prediction without retraining or auxiliary models.
Read the full abstract

All authors Raghavv Goel, Mukul Gagrani, Mingu Lee, Chris Lott

Computation and Language · Cross-listed from cs.LG

Weight Clustering Reveals the Importance of Relative Rank in Large Language Models

Borja Aizpurua, Sukhbinder Singh and Román Orús

Large language models (LLMs) contain billions of parameters, but many exact weight values are not essential. We apply scalar $K$-means clustering to pretrained models, replacing each weight matrix with only 16–64 shared values. For Llama~3.1-8B-Instruct and SmolLM2-135M, this preserves strong accuracy without retraining, providing a simple way to compress LLMs on disk. Fine-tuning only the cluster centroids further recovers 30–40% of the remaining accuracy loss. Using clustering as an analysis tool, we then keep the cluster assignments fixed and perturb only the shared values, to study which of their properties matter most. We find that preserving their relative rank (which connections are stronger or weaker than others) is more important than preserving their exact magnitudes. Breaking this ordering greatly increases perplexity, while changing the values but keeping the ordering causes little loss in many middle and late layers. However, rank preservation alone is not sufficient. Model performance also depends on maintaining the variance of the reconstructed weights, while early layers remain more sensitive; perturbing the cluster assignments themselves shows that their spatial geometry is also load-bearing. We observe the same behavior across four LLMs from two architecture families. These results suggest that the ordering of weights, together with their overall variance, plays a central role in LLM compression and robustness.
Read the full abstract

Computation and Language

IndicSafe: A Benchmark for Evaluating Multilingual LLM Safety in South Asia

Priyaranjan Pattnayak and Sanchari Chowdhuri

As large language models (LLMs) are deployed in multilingual settings, their safety behavior in culturally diverse, low-resource languages remains poorly understood. We present the first systematic evaluation of LLM safety across 12 Indic languages, spoken by over 1.2 billion people but underrepresented in LLM training data. Using a dataset of 6,000 culturally grounded prompts spanning caste, religion, gender, health, and politics, we assess 10 leading LLMs on translated variants of the prompt. Our analysis reveals significant safety drift: cross-language agreement is just 12.8%, and SAFE rate variance exceeds 17% across languages. Some models over-refuse benign prompts in low-resource scripts, overflag politically sensitive topics, while others fail to flag unsafe generations. We quantify these failures using prompt-level entropy, category bias scores, and multilingual consistency indices. Our findings highlight critical safety generalization gaps in multilingual LLMs and show that safety alignment does not transfer evenly across languages. We release IndicSafe, the first benchmark to enable culturally informed safety evaluation for Indic deployments, and advocate for language-aware alignment strategies grounded in regional harms.
Read the full abstract

Computation and Language

Pretrained Multilingual Transformers Reveal Quantitative Distance Between Human Languages

Yue Zhao, Jiatao Gu, Paloma Jeretič et al. (+1 more)

Understanding the distance between human languages is central to linguistics, anthropology, and tracing human evolutionary history. Yet, while linguistics has long provided rich qualitative accounts of cross-linguistic variation, a unified and scalable quantitative approach to measuring language distance remains lacking. In this paper, we introduce a method that leverages pretrained multilingual language models as systematic instruments for linguistic measurement. Specifically, we show that the spontaneously emerged attention mechanisms of these models provide a robust, tokenization-agnostic measure of cross-linguistic distance, termed Attention Transport Distance (ATD). By treating attention matrices as probability distributions and measuring their geometric divergence via optimal transport, we quantify the representational distance between languages during translation. Applying ATD to a large and diverse set of languages, we demonstrate that the resulting distances recover established linguistic groupings with high fidelity and reveal patterns aligned with geographic and contact-induced relationships. Furthermore, incorporating ATD as a regularizer improves transfer performance in low-resource machine translation. Our results establish a principled foundation for testing linguistic hypotheses using artificial neural networks. This framework transforms multilingual models into powerful tools for quantitative linguistic discovery, facilitating more equitable multilingual AI.
Read the full abstract

All authors Yue Zhao, Jiatao Gu, Paloma Jeretič, Weijie Su

Computation and Language

DebugLM: Learning Traceable Training Data Provenance for LLMs

Wenjie Jacky Mo, Qin Liu, Xiaofei Wen et al. (+3 more)

Large language models (LLMs) are trained through multi-stage pipelines over heterogeneous data sources, yet developers lack a principled way to pinpoint the specific data responsible for an observed behavior. This lack of observability reduces debugging to reactive patching and makes failures prone to recur under distribution shift or subsequent model updates. To address this limitation, we propose DebugLM, a framework that equips LLMs with built-in data provenance, enabling them to explicitly trace the origins of their behaviors to specific training data sources. Specifically, the model learns to associate its responses with unique provenance tags that indicate the responsible dataset, empowering developers to precisely identify where undesirable behaviors are learned. Building on this capability, DebugLM further supports targeted test-time remediation, enabling developers to selectively trigger targeted refusal for specified data sources without retraining or modifying model parameters. Experiments demonstrate that DebugLM provides accurate behavior tracing in multi-stage training pipelines and effective test-time remediation while preserving the general utility of the model.
Read the full abstract

All authors Wenjie Jacky Mo, Qin Liu, Xiaofei Wen, Wenxuan Zhou, Zhe Zhao, Muhao Chen

Computation and Language

Mitigating LLM Hallucinations through Domain-Grounded Tiered Retrieval

Md. Asraful Haque, Aasar Mehdi, Maaz Mahboob et al. (+1 more)

Large Language Models (LLMs) have achieved unprecedented fluency but remain susceptible to "hallucinations" - the generation of factually incorrect or ungrounded content. This limitation is particularly critical in high-stakes domains where reliability is paramount. We propose a domain-grounded tiered retrieval and verification architecture designed to systematically intercept factual inaccuracies by shifting LLMs from stochastic pattern-matchers to verified truth-seekers. The proposed framework utilizes a four-phase, self-regulating pipeline implemented via LangGraph: (I) Intrinsic Verification with Early-Exit logic to optimize compute, (II) Adaptive Search Routing utilizing a Domain Detector to target subject-specific archives, (III) Refined Context Filtering (RCF) to eliminate non-essential or distracting information, and (IV) Extrinsic Regeneration followed by atomic claim-level verification. The system was evaluated across 650 queries from five diverse benchmarks: TimeQA v2, FreshQA v2, HaluEval General, MMLU Global Facts, and TruthfulQA. Empirical results demonstrate that the pipeline consistently outperforms zero-shot baselines across all environments. Win rates peaked at 83.7% in TimeQA v2 and 78.0% in MMLU Global Facts, confirming high efficacy in domains requiring granular temporal and numerical precision. Groundedness scores remained robustly stable between 78.8% and 86.4% across factual-answer rows. While the architecture provides a robust fail-safe for misinformation, a persistent failure mode of "False-Premise Overclaiming" was identified. These findings provide a detailed empirical characterization of multi-stage RAG behavior and suggest that future work should prioritize pre-retrieval "answerability" nodes to further bridge the reliability gap in conversational AI.
Read the full abstract

All authors Md. Asraful Haque, Aasar Mehdi, Maaz Mahboob, Tamkeen Fatima

Computation and Language

How do LLMs Compute Verbal Confidence

Dharshan Kumaran, Arthur Conmy, Federico Barbero et al. (+3 more)

Verbal confidence – prompting LLMs to state their confidence as a number or category – is widely used to extract uncertainty estimates from black-box models. However, how LLMs internally generate such scores remains unknown. We address two questions: first, when confidence is computed – just-in-time when requested, or automatically during answer generation and cached for later retrieval; and second, what verbal confidence represents – token log-probabilities, or a richer evaluation of answer quality? Focusing on Gemma 3 27B (across TriviaQA, BigMath, and MMLU), Qwen 2.5 7B, and the reasoning model Magistral Small 24B, we provide convergent evidence for cached retrieval. Activation steering, patching, noising, and swap experiments reveal that confidence representations emerge at answer-adjacent positions before appearing at the verbalization site. Attention blocking pinpoints the information flow: confidence is gathered from answer tokens, cached at the first post-answer position, then retrieved for output. Critically, linear probing and variance partitioning reveal that these cached representations explain substantial variance in verbal confidence beyond token log-probabilities, suggesting a richer answer-quality evaluation rather than a simple fluency readout. These findings demonstrate that verbal confidence reflects automatic, sophisticated self-evaluation – not post-hoc reconstruction – with implications for understanding metacognition in LLMs and improving calibration.
Read the full abstract

All authors Dharshan Kumaran, Arthur Conmy, Federico Barbero, Simon Osindero, Viorica Patraucean, Petar Veličković

Computation and Language

Event-Centric Human Value Understanding in News-Domain Texts: An Actor-Conditioned, Multi-Granularity Benchmark

Yao Wang, Xin Liu, Zhuochen Liu et al. (+5 more)

Existing human value datasets do not directly support value understanding in factual news: many are actor-agnostic, rely on isolated utterances or synthetic scenarios, and lack explicit event structure or value direction. We present NEVU (News Event-centric Value Understanding), a benchmark for actor-conditioned, event-centric, and direction-aware human value recognition in factual news. NEVU evaluates whether models can identify value cues, attribute them to the correct actor, and determine value direction from grounded evidence. Built from 2{,}865 English news articles, NEVU organizes annotations at four semantic unit levels (Subevent, behavior-based composite event, story-based composite event, and Article) and labels \mbox{(unit, actor)} pairs for fine-grained evaluation across local and composite contexts. The annotations are produced through an LLM-assisted pipeline with staged verification and targeted human auditing. Using a hierarchical value space with 54 fine-grained values and 20 coarse-grained categories, NEVU covers 45{,}793 unit–actor pairs and 168{,}061 directed value instances. We provide unified baselines for proprietary and open-source LLMs, and find that lightweight adaptation (LoRA) consistently improves open-source models, showing that although NEVU is designed primarily as a benchmark, it also supports supervised adaptation beyond prompting-only evaluation. Data availability is described in Appendix~\ref{app:data_code_availability}.
Read the full abstract

All authors Yao Wang, Xin Liu, Zhuochen Liu, Jiankang Chen, Adam Jatowt, Kyoungsook Kim, Noriko Kando, Haitao Yu

Computation and Language · Cross-listed from eess.AS

The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning

Donghang Wu, Tianyu Zhang, Yuxin Li et al. (+4 more)

During conversational interactions, humans subconsciously engage in concurrent thinking while listening to a speaker. Although this internal cognitive processing may not always manifest as explicit linguistic structures, it is instrumental in formulating high-quality responses. Inspired by this cognitive phenomenon, we propose a novel Full-duplex LAtent and Internal Reasoning method named FLAIR that conducts latent thinking simultaneously with speech perception. Unlike conventional "thinking" mechanisms in NLP, which require post-hoc generation, our approach aligns seamlessly with spoken dialogue systems: during the user's speaking phase, it recursively feeds the latent embedding output from the previous step into the next step, enabling continuous reasoning that strictly adheres to causality without introducing additional latency. To enable this latent reasoning, we design an Evidence Lower Bound-based objective that supports efficient supervised finetuning via teacher forcing, circumventing the need for explicit reasoning annotations. Experiments demonstrate the effectiveness of this think-while-listening design, which achieves competitive results on a range of speech benchmarks. Furthermore, FLAIR robustly handles conversational dynamics and attains competitive performance on full-duplex interaction metrics.
Read the full abstract

All authors Donghang Wu, Tianyu Zhang, Yuxin Li, Hexin Liu, Chen Chen, Eng Siong Chng, Yoshua Bengio

Computation and Language

Text-to-Stage: Spatial Layouts from Long-form Narratives

Jefferson Hernandez, Swarnadeep Saha, Chenxi Whitehouse et al. (+6 more)

In this work, we probe the ability of a language model to demonstrate spatial reasoning from unstructured text, mimicking human capabilities and automating a process that benefits many downstream media applications. Concretely, we study the narrative-to-play task: inferring stage-play layouts (scenes, speaker positions, movements, and room types) from text that lacks explicit spatial, positional, or relational cues. We then introduce a dramaturgy-inspired deterministic evaluation suite and, finally, a training and inference recipe that combines rejection SFT using Best-of-N sampling with RL from verifiable rewards via GRPO. Experiments on a text-only corpus of classical English literature demonstrate improvements over vanilla models across multiple metrics (character attribution, spatial plausibility, and movement economy), as well as alignment with an LLM-as-a-judge and subjective human preferences.
Read the full abstract

All authors Jefferson Hernandez, Swarnadeep Saha, Chenxi Whitehouse, Sanjeel Parekh, Calvin Murdock, Yuliang Li, W. Owen Brimijoin, Vamsi Krishna Ithapu, Ishwarya Ananthabhotla

Computation and Language · Cross-listed from cs.SE

CodeScout: An Effective Recipe for Reinforcement Learning of Code Search Agents

Lintang Sutawika, Aditya Bharat Soni, Bharath Sriraam R R et al. (+8 more)

A prerequisite for coding agents to perform tasks on large repositories is code localization - the identification of relevant files, classes, and functions to work on. While repository-level code localization has been performed using embedding-based retrieval approaches such as vector search, recent work has focused on developing agents to localize relevant code either as a standalone precursor to or interleaved with performing actual work. Most prior methods on agentic code search equip the agent with complex, specialized tools, such as repository graphs derived from static analysis. In this paper, we demonstrate that, with an effective reinforcement learning recipe, a coding agent equipped with nothing more than a standard Unix terminal can be trained to achieve strong results. Our experiments on three benchmarks (SWE-Bench Verified, Pro, and Lite) reveal that our models consistently achieve superior or competitive performance over 2-18x larger base and post-trained LLMs and sometimes approach performance provided by closed models like Claude Sonnet, even when using specialized scaffolds. Our work particularly focuses on techniques for re-purposing existing coding agent environments for code search, reward design, and RL optimization. We release the resulting model family, CodeScout, along with all our code and data for the community to build upon.
Read the full abstract

All authors Lintang Sutawika, Aditya Bharat Soni, Bharath Sriraam R R, Apurva Gandhi, Taha Yassine, Sanidhya Vijayvargiya, Yuchen Li, Xuhui Zhou, Yilin Zhang, Leander Melroy Maben, Graham Neubig

Computation and Language · Cross-listed from cs.LG

Discovering Decoupled Functional Modules in Large Language Models

Yanke Yu, Jin Li, Ying Sun et al. (+3 more)

Understanding the internal functional organization of Large Language Models (LLMs) is crucial for improving their trustworthiness and performance. However, how LLMs organize different functions into modules remains highly unexplored. To bridge this gap, we formulate a functional module discovery problem and propose an Unsupervised LLM Cross-layer MOdule Discovery (ULCMOD) framework that simultaneously disentangles the large set of neurons in the entire LLM into modules while discovering the topics of input samples related to these modules. Our framework introduces a novel objective function and an efficient Iterative Decoupling (IterD) algorithm. Extensive experiments show that our method discovers high-quality, disentangled modules that capture more meaningful semantic information and achieve superior performance in various downstream tasks. Moreover, our qualitative analysis reveals that the discovered modules show semantic coherence, correspond to interpretable specializations, and a clear spatial and hierarchical organization within the LLM. Our work provides a novel tool for interpreting the functional modules of LLMs, filling a critical blank in LLM's interpretability research.
Read the full abstract

All authors Yanke Yu, Jin Li, Ying Sun, Ping Li, Zhefeng Wang, Yi Zheng

Computation and Language · Cross-listed from eess.AS

Multi-Source Evidence Fusion for Audio Question Answering

Aivo Olev and Tanel Alumäe

Large audio language models (LALMs) can answer questions about speech, music, and environmental sounds, yet their internal reasoning is largely opaque and difficult to validate. We describe TalTech's solution to the Agent Track of the Interspeech 2026 Audio Reasoning Challenge, in which systems are evaluated on reasoning process quality, specifically the factual accuracy, logical soundness, and completeness of their reasoning chains. Our multi-source ensemble pipeline uses two LALMs that generate independent observations, while a separate text-only reasoning model cross-checks these against outputs from 25 acoustic tools organized into reliability tiers. By grounding every inference step in explicit, reliability-tagged evidence, the system produces dense, verifiable reasoning chains. Our system ranked first in the challenge, outperforming all competing systems by a wide margin in challenge's reasoning quality metric.
Read the full abstract

Computation and Language

RAWR: Reward Assignment Without Rollouts in Verifiable Domains

Corentin Royer, Anna Hedström, Debarun Bhattacharjya et al. (+3 more)

Understanding and evaluating multi-step reasoning in LLMs at the level of individual steps remains a key challenge. Process reward models (PRMs) provide a solution by scoring each step, enabling fine-grained supervision and improved reliability. However, training them requires costly human annotation or computationally intensive rollout-based labeling. To solve this, we introduce MCNIG, a scalable method for automatically labeling the quality of individual reasoning steps in any verifiable domain. Its step score, net information gain (NetIG), improves upon single-reference information gain (IG) by comparing the most-supported correct answer against the most-supported incorrect one, yielding a robust signal even for long and structured outputs like code and SQL, where IG fails. We show that the signal produced by MCNIG correlates with human judgments of step quality, and we apply MCNIG labels to train PRMs that achieve the best average best-of-K accuracy across eight benchmarks spanning mathematics, code generation, text-to-SQL, and scientific QA. Crucially, MCNIG generates no rollouts, cutting labeling complexity to O(N) and making it up to X times cheaper than rollout-based methods at comparable label quality, which makes large-scale process supervision practical.
Read the full abstract

All authors Corentin Royer, Anna Hedström, Debarun Bhattacharjya, Gaetano Rossiello, Andrea Giovannini, Mennatallah El-Assady

Computation and Language · Cross-listed from cs.AI

Governed Memory: A Production Architecture for Multi-Agent Workflows

Hamed Taheri

Enterprise AI deploys dozens of autonomous agent nodes across workflows, each acting on the same entities with no shared memory and no common governance. We identify five structural challenges arising from this memory governance gap: memory silos across agent workflows; governance fragmentation across teams and tools; unstructured memories unusable by downstream systems; redundant context delivery in autonomous multi-step executions; and silent quality degradation without feedback loops. We present Governed Memory, a shared memory and governance layer addressing this gap through four mechanisms: a dual memory model combining open-set atomic facts with schema-enforced typed properties; tiered governance routing with progressive context delivery; reflection-bounded retrieval with entity-scoped isolation; and a closed-loop schema lifecycle with AI-assisted authoring and automated per-property refinement. We validate each mechanism through controlled experiments (N=250, five content types): 99.6% fact recall with complementary dual-modality coverage; 92% governance routing precision; 50% token reduction from progressive delivery; zero cross-entity leakage across 500 adversarial queries; 100% adversarial governance compliance; and output quality saturation at approximately seven governed memories per entity. On the LoCoMo benchmark, the architecture achieves 74.8% overall accuracy, confirming that governance and schema enforcement impose no retrieval quality penalty. The system is in production at Personize.ai.
Read the full abstract

Computation and Language

CoVerRL: Breaking the Consensus Trap in Label-Free Reasoning via Generator-Verifier Co-Evolution

Teng Pan, Yuchen Yan, Zixuan Wang et al. (+6 more)

Label-free reinforcement learning enables large language models to improve reasoning capabilities without ground-truth supervision, typically by treating majority-voted answers as pseudo-labels. However, we identify a critical failure mode: as training maximizes self-consistency, output diversity collapses, causing the model to confidently reinforce systematic errors that evade detection. We term this the consensus trap. To escape it, we propose CoVerRL, a framework where a single model alternates between generator and verifier roles, with each capability bootstrapping the other. Majority voting provides noisy but informative supervision for training the verifier, while the improving verifier progressively filters self-consistent errors from pseudo-labels. This co-evolution creates a virtuous cycle that maintains high reward accuracy throughout training. Experiments across Qwen and Llama model families demonstrate that CoVerRL outperforms label-free baselines by 4.7-5.9% on mathematical reasoning benchmarks. Moreover, self-verification accuracy improves from around 55% to over 85%, confirming that both capabilities genuinely co-evolve.
Read the full abstract

All authors Teng Pan, Yuchen Yan, Zixuan Wang, Ruiqing Zhang, Guiyang Hou, Wenqi Zhang, Weiming Lu, Jun Xiao, Yongliang Shen

Computation and Language · Cross-listed from cs.SD

Modeling Overlapped Speech with Shuffles

Matthew Wiesner, Samuele Cornell, Alexander Polok et al. (+3 more)

We propose to model parallel streams of data, such as overlapped speech, using shuffles. Specifically, this paper shows how the shuffle product and partial order finite-state automata (FSAs) can be used for alignment and speaker-attributed transcription of overlapped speech. We train using the total score on these FSAs as a loss function, marginalizing over all possible serializations of overlapping sequences at subword, word, and phrase levels. To reduce graph size, we impose temporal constraints by constructing partial order FSAs. We address speaker attribution by modeling (token, speaker) tuples directly. Viterbi alignment through the shuffle product FSA directly enables one-pass alignment. We evaluate performance on synthetic LibriSpeech overlaps. To our knowledge, this is the first algorithm that enables single-pass alignment of multi-talker recordings. All algorithms are implemented using k2 / Icefall.
Read the full abstract

All authors Matthew Wiesner, Samuele Cornell, Alexander Polok, Lucas Ondel Yang, Lukáš Burget, Sanjeev Khudanpur

Computation and Language

Harm or Humor: A Multimodal, Multilingual Benchmark for Overt and Covert Harmful Humor

Ahmed Sharshar, Hosam Elgendy, Saad El Dine Ahmed et al. (+2 more)

Dark humor often relies on subtle cultural nuances and implicit cues that require contextual reasoning to interpret, posing safety challenges that current static benchmarks fail to capture. To address this, we introduce a novel multimodal, multilingual benchmark for detecting and understanding harmful and offensive humor. Our manually curated dataset comprises 3,000 texts and 6,000 images in English and Arabic, alongside 1,200 videos that span English, Arabic, and language-independent (universal) contexts. Unlike standard toxicity datasets, we enforce a strict annotation guideline: distinguishing Safe jokes from Harmful ones, with the latter further classified into Explicit (overt) and Implicit (Covert) categories to probe deep reasoning. We systematically evaluate state-of-the-art (SOTA) open and closed-source models across all modalities. Our findings reveal that closed-source models significantly outperform open-source ones, with a notable difference in performance between the English and Arabic languages in both, underscoring the critical need for culturally grounded, reasoning-aware safety alignment. Warning: this paper contains example data that may be offensive, harmful, or biased.
Read the full abstract

All authors Ahmed Sharshar, Hosam Elgendy, Saad El Dine Ahmed, Yasser Rohaim, Yuxia Wang

Computation and Language

Adaptive Guidance for Retrieval-Augmented Masked Diffusion Models

Jaemin Kim and Jong Chul Ye

Retrieval-Augmented Generation (RAG) improves factual grounding by incorporating external knowledge into language model generation. However, when retrieved context is noisy, unreliable, or inconsistent with the model's parametric knowledge, it introduces retrieval-prior conflicts that can degrade generation quality. While this problem has been studied in autoregressive language models, it remains largely unexplored in diffusion-based language models, where the iterative denoising process introduces unique challenges for integrating retrieved context. In this work, we propose Adaptive Retrieval-Augmented Masked Diffusion (ARAM), a training-free adaptive guidance framework for Masked Diffusion Models (MDMs) in RAG settings. ARAM dynamically calibrates the guidance scale during denoising according to the Signal-to-Noise Ratio (SNR) of the distributional shift induced by retrieved context. Intuitively, the model strengthens guidance when the retrieved context provides reliable corrective evidence and suppresses it when the contextual signal is noisy or non-supportive. Extensive experiments on multiple knowledge-intensive QA benchmarks show that ARAM improves overall QA performance over competitive RAG baselines.
Read the full abstract