Peer Revenue Fully funded

Papers

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

Computer Science Computation and Language

cs.CL Watch Computation and Language

Tuesday 22 September

Computation and Language

A Holistic Assessment of the Carbon Footprint of Noor, a Very Large Arabic Language Model

Imad Lakim, Ebtesam Almazrouei, Ibrahim Abu Alhaol et al. (+2 more)

As ever larger language models grow more ubiquitous, it is crucial to consider their environmental impact. Characterised by extreme size and resource use, recent generations of models have been criticised for their voracious appetite for compute, and thus significant carbon footprint. Although reporting of carbon impact has grown more common in machine learning papers, this reporting is usually limited to compute resources used strictly for training. In this work, we propose a holistic assessment of the footprint of an extreme-scale language model, Noor. Noor is an ongoing project aiming to develop the largest multi-task Arabic language models – with up to 13B parameters – leveraging zero-shot generalisation to enable a wide range of downstream tasks via natural language instructions. We assess the total carbon bill of the entire project: starting with data collection and storage costs, including research and development budgets, pretraining costs, future serving estimates, and other exogenous costs necessary for this international cooperation. Notably, we find that inference costs and exogenous factors can have a significant impact on total budget. Finally, we discuss pathways to reduce the carbon footprint of extreme-scale models.
Read the full abstract

All authors Imad Lakim, Ebtesam Almazrouei, Ibrahim Abu Alhaol, Merouane Debbah, Julien Launay

Computation and Language

Automatic estimation of verbal fluency index in people with Motor Neuron Disease using ASR alignment and pause modelling

Bahman Mirheidari, Leslie Ing, Daniel Blackburn et al. (+3 more)

Monitoring cognitive impairment (CI) in motor neuron disease (MND) is essential for timely treatment and care, yet challenging due to co-occurring speech difficulties. The Edinburgh Cognitive and Behavioural ALS Screen (ECAS) provides a robust metric for CI assessment, with the Verbal Fluency Index (VFI) a central element. Building on recent advances in automated speech analysis, this study proposes a system for estimating VFI. It leverages a unique MND dataset and combines ASR (WhisperX) and VAD (Silero) with refined timestamping to predict the VFI and extract several clinically interpretable measures. Our approach outperformed systems based on traditional acoustic features and self-supervised embeddings, evaluated using multiple regression algorithms. Clinically inspired features consistently outperformed the other sets, with the best models achieving strong results (P-words: R2 0.9, NRMSE 0.05; S-words: R2 0.8, NRMSE 0.08), demonstrating the feasibility of automated VFI estimation.
Read the full abstract

All authors Bahman Mirheidari, Leslie Ing, Daniel Blackburn, Sharon Abrahams, Christopher McDermott, Heidi Christensen

Computation and Language

TomasuLLM: Out-of-Order Speculative Execution for LLM Agents

Jiangnan Yu, Ceyu Xu, Mengming Li et al. (+6 more)

Long-running tools can dominate coding-agent latency: compilers, test suites, and repository commands take seconds to minutes while the agent idles. This observation stall presents the same tension that drove out-of-order processors – asequential interface hides work that can be predicted and started early, but a speculative result may become visible only after it and every earlier step have been validated. We present TomasuLLM, a runtime that executes agent tool calls out of trajectory order while preserving task-execution correctness. It drafts future actions, runs them in isolated copy-on-write sandboxes, traces their dependencies and effects, and commits results in trajectory order only after validation against committed state. Across three benchmarks spanning sub-second to minutes-long tool calls, TomasuLLM improves the reported benchmark means and scales with tool latency: 1.31x on 100 SWE-bench Verified tasks, 1.35x on 28 Terminal-Bench 2.0 tasks, and 1.27x matched progress on 18 SWE-Marathon sessions. Across 4,010 audited commit-validation records, it produces zero false accepts.
Read the full abstract

All authors Jiangnan Yu, Ceyu Xu, Mengming Li, Shiyu Huang, Yiran Xia, Jian Weng, Hui Xue, Haohui Mai, Yuan Xie

Computation and Language

Reliable but Design-Sensitive: Instrument Uncertainty in LLM Annotation

Thomas Reiter, Christoph Kern, Fedor Miasnikov et al. (+4 more)

Large language models (LLMs) can give reliable labels under one setup yet change those labels when researchers make other reasonable design choices. We tested seven LLMs, 12 task designs, three independent runs, and 3,000 tweets labeled for offensive language and hate speech. Repeating the same model and task design produced high agreement (median Fleiss' $\kappa = 0.91$). Agreement fell when we changed the task design for the same tweets (median Cohen's $\kappa = 0.76$). Task design and model choice increased the variance of estimated prevalence by factors of 76.7 for offensive language and 110.6 for hate speech compared with sampling variance alone. Variation across LLM task designs reached 560-572 basis points, compared with 270-331 basis points across five human instrument versions. Confidence scores did not solve this problem. They tracked repeated model outputs more closely than agreement with human labels, and grouping six tweets in one prompt lowered mean offensive-language confidence by 660 basis points. We call the variation caused by task design and model choice instrument uncertainty. Researchers can measure it only by comparing reasonable task designs. Repeating one setup or relying on confidence scores cannot replace that test.
Read the full abstract

All authors Thomas Reiter, Christoph Kern, Fedor Miasnikov, Sofiia Nikolenko, Rob Chew, Stephanie Eckman, Frauke Kreuter

Computation and Language

The Ongiini-Eval-OW Benchmark: A Concept Paper for the Planned Benchmarking of Machine Translation and Large Language Models on Oshindonga and Oshikwanyama

Sebastian Küpers

Oshiwambo – a cluster of mutually intelligible Bantu languages spoken by over a million people across northern Namibia and southern Angola, and the home language of roughly half of Namibian households – has, to our knowledge, no published machine-translation evaluation benchmark. Major commercial services (Google Translate, DeepL, Microsoft Translator), open multilingual MT models (NLLB-200, MADLAD-400), and the open Masakhane checkpoint collection all lack coverage of either standardised dialect, Oshindonga or Oshikwanyama. We announce Ongiini-Eval-OW, a planned 600-item English-Oshindonga and English-Oshikwanyama benchmark with native-speaker references from two independent translators, a 30-item inter-translator agreement set, an 11-tag phenomenon-tagged stratification (at least 30 items per tag), a deterministic 30% blind split, and a reproducible scoring protocol over chrF++, BLEU, and COMET-22, supplemented by a 50-item human-evaluation round. We document the empirical coverage gap, the dataset composition, the launch-leaderboard model matrix across American, European, and Chinese frontier and open-weight systems, and the contribution pipeline. The dataset is targeted for first public release in Q4 2026; this v1.0 concept paper announces the design and the call for participation. Data and code will be released under CC-BY-4.0 and MIT respectively.
Read the full abstract

Computation and Language

Count Evidence, Not Sentences: Tempered Evidence Fusion of LLM Judgments for Long-Text Value Measurement

Yuhe Wu, Rui Qian, Guangyu Wang et al. (+10 more)

Large language models (LLMs) are increasingly used to measure public value orientations from long social media posts, yet such posts often mix background, quotations, concessions, and only a few stance-bearing sentences. Existing approaches either ask the model to predict a document-level label directly, which can be overconfident, or aggregate sentence-level predictions by majority or soft voting, which treat uncertain and decisive sentences as equally informative. We formulate long-text value measurement as a decision-fusion problem and propose Tempered Evidence Fusion (TEF), a training-free rule that weights each sentence's log-odds by its normalized information gain, as derived from a generalized Bayesian posterior. This makes the fused score nearly vanish for uncertain sentences while preserving the Bayes-optimal weight of decisive evidence. We further introduce Multi-event Insight Network Dimensions (MIND), a benchmark of 8,358 Chinese and English posts spanning five years of public events and six value dimensions. On MIND, TEF outperforms the strongest baseline among Direct, Majority Vote, and Soft Vote by an average of 4.5 accuracy points and 4.6 macro-F1 points across five LLMs and two languages. MIND dataset and code are available at https://github.com/Kzczc/ICASSP2027-TEF.
Read the full abstract

All authors Yuhe Wu, Rui Qian, Guangyu Wang, Yuran Chen, Yuanchao Zhu, Junjie Yang, Zhengheng Li, Jiulin Cai, Tianyi Zhang, Zihan Dong, Jiaxin Liu, Yujie Chen, Guang Zhang

Computation and Language · Cross-listed from cs.LG

The Linear Representation Hypothesis Needs a Group Action

Louie Hong Yao, Yuhao Li and Shengchao Liu

To make claims about representations that generalize beyond a particular trained model, we need to specify when two representations should count as equivalent. The Linear Representation Hypothesis is often discussed without making this equivalence explicit. Different notions of equivalence preserve different structures, so metrics, probes, and interventions that appear to study the same representation may in fact correspond to different hypotheses. We therefore argue that the Linear Representation Hypothesis is not one hypothesis but a family of claims distinguished by representation equivalence. We formalize this idea using group actions, specifying the representation object, the procedure that produces it, and the property ultimately asserted, while accounting for equivalences imposed by the model architecture. This framework clarifies how assumptions can change across metrics, reading points, and analysis stages, and we use it to audit common representation quantities and recent interpretability analyses.
Read the full abstract

Computation and Language

Giving Credit Where It's Due: Redundancy-Aware Learning for Efficient Reasoning

Yuqing Zhou, Hong Wang, Manqing Mao et al. (+8 more)

Large reasoning models can produce correct yet unnecessarily long reasoning traces. Existing methods improve reasoning efficiency with trajectory-level objectives or local token- and step-level signals, but rarely model inter-step semantic dependencies. This limits their ability to distinguish redundant steps from those that support later deductions, making it harder to shorten reasoning without sacrificing accuracy. We introduce RECAP (REdundancy-aware Credit Assignment via Propagation), which addresses this limitation by assigning credit where it is due based on both a step's downstream role in the reasoning structure and its contribution to solving the problem correctly. We define structural responsibility to capture the step's downstream role by measuring how strongly later reasoning depends on it, using credit propagated backward from the final-answer node through an outcome-independent, LLM-annotated semantic dependency graph. However, a step can have high structural responsibility yet steer the reasoning away from the correct solution. RECAP therefore introduces step efficacy to measure answer-directed progress through changes in gold-answer log-likelihood as each step is added. Together, these signals reshape rollout-level GRPO advantages into step-specific updates. RECAP requires neither a separately trained process reward model nor preconstructed concise trajectories. Across two 7B models and four mathematical reasoning benchmarks, RECAP improves the accuracy-efficiency trade-off. On Qwen2.5-Math-7B, it improves pass@1 by 2.0-3.7 percentage points while reducing reasoning tokens by 8%-31% relative to GRPO across all four benchmarks. Analysis suggests these savings reflect fewer reasoning operations and less dead-end reasoning, rather than more compact expression.
Read the full abstract

All authors Yuqing Zhou, Hong Wang, Manqing Mao, Zhuoer Wang, Samson Koelle, Jie Yuan, Yanjun Lin, James Feng, Nikki Lijing Kuang, Ziwei Zhu, Wei Niu

Computation and Language · Cross-listed from cs.CV

Feed the Panel Dimensions, Not Verdicts: Rubric-Decomposed Fusion of Vision-Language Aesthetic Judges

Amit Jadhav, Shaurya Beriwala and Beomjin Kim

Vision-language models (VLMs) are deployed as zero-shot judges of image aesthetics, and panels of several models are recommended, on thin evidence, as the way to make such judges reliable. On two human-rated datasets, EVA and PARA, we find that a panel of holistic judges never significantly beats its best member, whether the verdicts are averaged or fused by a learned combiner. What a panel is worth depends on what it is fed. We therefore have each model score each image on the five dimensions of a frozen, human-written rubric and fuse those scores, alongside each model's verdict, across model families with an out-of-fold combiner. The dimension scores measure what their labels claim: with the overall human score partialled out, a dimension prompt carries more attribute-specific information than the holistic prompt in 28 of 30 model-attribute cells. Fused, they beat the best single VLM in all ten three-family panels on EVA (against that best single model, +0.07 Spearman rho for the strongest trio and +0.10 for the pre-declared one, and +0.06 and +0.07 when averaged over twenty fold partitions; against the panel mean, the primary test gives +0.118 on its EVA design set), and on PARA they reach parity under Spearman rho and a small, non-significant loss under Kendall tau-b, where one model already captures 85% of the human noise ceiling. It is not a feature-count artefact: giving the same combiner an equal number of pure holistic columns, split from the same repetitions, does not reproduce it. The gain costs a few hundred labels, which do not transfer between datasets, and 4.8x the API calls on EVA; we report it with paired bootstraps and Kendall tau-b, alongside a failed pre-registration and the configurations that lost.
Read the full abstract

Computation and Language

NADI 2026: The Second Multidialectal Arabic Speech Processing Shared Task

Peter Sullivan, Bashar Talafha, Ahmed Ashraf et al. (+11 more)

NADI 2026 is the seventh edition of the Nuanced Arabic Dialect Identification (NADI) shared task series and the second dedicated to multidialectal Arabic speech processing. This edition comprises five tasks and eight subtasks spanning Automatic Speech Recognition (ASR), Spoken Dialect Identification (SDID), Text-to-Speech (TTS), Spoken Language Translation (SLT), and Spoken Language Understanding (SLU). NADI 2026 emphasizes realistic evaluation through low-bandwidth, mixed-dialect, code-switched, out-of-domain, and zero-shot settings, while introducing TTS, SLT, and SLU to the series for the first time. The shared task attracted 21 participating teams from at least 13 countries, with 48 test-phase submissions and 14 submitted system-description papers. Results show that out-of-domain generalization remains a major bottleneck and highlight the effectiveness of recent Arabic-specialized speech models, multimodal dialect identification approaches, and ensemble methods. Overall, NADI 2026 provides a broader and more challenging benchmark for robust Arabic dialect speech processing.
Read the full abstract

All authors Peter Sullivan, Bashar Talafha, Ahmed Ashraf, Fethi Bougares, Haroun Elleuch, Chiyu Zhang, AbdelRahim Elmadany, Youssef Mohamed, Salima Mdhaffar, Yannick Estève, Mohamed Elhoseiny, Hamzah Luqman, Nizar Habash, Muhammad Abdul-Mageed

Computation and Language · Cross-listed from cs.LG

ChipMEM: Verification-Grounded Memory for EDA Agents

Abdulrahman AlRabah, Joshua Mabry, Dilek Hakkani-Tür et al. (+4 more)

Large language model (LLM)-based agents use Electronic Design Automation (EDA) tools to generate and revise register-transfer-level (RTL) designs under synthesis and verification feedback. Recent methods learn from this feedback by distilling reusable skills from execution traces or by training on rewards derived from EDA-tools. Both methods are typically evaluated on the tasks that produced the experience. Repeated access to benchmark feedback on the same task can reward task-specific revision rather than creating reusable knowledge that transfers. We introduce ChipMEM, a verification-grounded memory layer for EDA agents. It combines cross-task procedural memory with within-trajectory statistical guidance. Its procedural component distills and stores a skill only after it passes synthesis, simulation, or formal checks, rather than relying on model self-assessments. A Bayesian component maintains hierarchical Beta estimates over tool-call outcomes and ranks recovery strategies that succeeded under comparable errors. A common adapter applies the same memory interface to RTL optimization and testbench-generation agents while preserving each domain's tools and acceptance criteria. We measure performance on training tasks and evaluate whether learned skills transfer to unseen tasks. On RTLRewriter-Bench, under matched model and tool settings, ChipMEM produces equivalence-passing outputs on 39/54 scored designs versus 35/54 without memory; on the 49-design short suite, mean area improvement is 8.69% versus 5.66%. On held-out CVDP tasks, ChipMEM with a frozen procedural library achieves 20/20 accepted outcomes versus 18/20 without memory in a single evaluation per setting.
Read the full abstract

All authors Abdulrahman AlRabah, Joshua Mabry, Dilek Hakkani-Tür, Abdussalam Alawini, Hamid Shojaei, Kartik Hegde, Sandesh Adhikary

Computation and Language

What Changes When Fact-Verification Scores Improve? Evidence and Answer Accounting Across Trained Verifiers and LLMs

Han Chen and Yingrui Li

A joint fact-verification score assesses answers and submitted evidence together. When the score improves, how much of the gain remains if the answers are held fixed? On FEVEROUS, strict score is the percentage of claims with a correct answer and a complete annotated evidence group in the submitted evidence. Across four trained DeBERTa checkpoints and 7,890 claims, replacing DCUF evidence with UnifEE evidence raises strict score by 9.61 percentage points, compared with 1.96 percentage points in answer accuracy. The paired 95% interval for the strict-score gain is [8.77, 10.43], conditional on these checkpoints. Replacing only the evidence passed to the scorer accounts for 7.92 or 9.08 percentage points when we retain the answers generated from DCUF or UnifEE evidence, respectively. To examine how this evidence gain depends on evaluation choices, we generate 470,400 responses from two 8B LLMs on FEVER, FEVEROUS, and SciFact under two answer formats and two context budgets. Increasing context from 256 to 2,048 tokens raises the fixed-answer evidence gain on FEVEROUS by 3.84 and 3.10 percentage points for Qwen and Llama, respectively. The effects fall short of the prespecified cross-dataset criterion, while some intervals extend beyond the two-point small-effect bound. Post-hoc analyses quantify changes in answers and submitted evidence, and show when aggregate accuracy and evidence-coverage rates miss the claim-level pattern. The four answer-evidence score combinations reveal changes that endpoint and aggregate metrics leave unresolved.
Read the full abstract

Computation and Language

The Illinois Social Attitudes Aggregate Corpus (ISAAC): An Open Tool and Reproducible Pipeline for Analyzing Social Group Discourse at Scale

Babak Hemmatian, Sarah Hadjarab, Jessica Chen et al. (+1 more)

We introduce the Illinois Social Attitudes Aggregate Corpus (ISAAC), an open, modular, and accessible corpus of 527 million+ English-language Reddit posts selected for relevance to six key social group distinctions based on race, sexuality, age, ability, body weight, and skin tone, covering the 17-year period from 2007 to 2023. A multi-step, human-audited filtering pipeline was used to keep irrelevant content in the curated dataset below 10%, both overall and for each social group distinction. Each post was then algorithmically annotated with the user's estimated home region, along with a suite of validated off-the-shelf and custom semantic labels including moralization, sentiment, emotion, and linguistic generalization. We confirm the validity of the resulting corpus through convergent evidence linking ISAAC to macro-level societal trends, such as online search behavior, temporal spikes during major societal events (both nationally and regionally), and long-term shifts in public attitudes. By offering a unified, public infrastructure, ISAAC eliminates research fragmentation and enables seamless replication while supporting diverse empirical workflows at scale. Specifically, ISAAC allows investigators to perform cross-category comparisons, conduct high-precision tracking of long-term temporal shifts in social group discourse, and map spatial variation onto localized public opinion and policy outcomes. ISAAC's fully public, modular pipeline facilitates easy extension of the corpus to new platforms, languages, and social categories. To accommodate various research needs, ISAAC is accessible both without coding through a point-and-click website and labeler web-apps, and programmatically via an SQL playground, a Python package, and HuggingFace.
Read the full abstract

All authors Babak Hemmatian, Sarah Hadjarab, Jessica Chen, Benedek Kurdi

Computation and Language

EduBehaviors: Assertion-based Schemas for Auditable Coding of Educational Dialogues

Julian Bernado, Ana Trindade Ribeiro, Xander Beberman et al. (+1 more)

Large language models have enabled rapid deployment of language classification and analysis, allowing users to describe complex constructs in natural language, such as pedagogical moves or expressions of learning and engagement, and label them in conversational datasets such as classroom or tutoring transcripts. However, because LLM reasoning is opaque, we have no verifiable, mechanistic insight into why a model chose a label for an utterance, and output differences over time or across models challenge replication. We introduce the EduBehaviors framework to generate interpretable, reliable, and scalable construct annotations in educational dialogue data that retain human agency while leveraging the strengths of generative AI, encoders, and machine learning models. We evaluate the framework on the TalkMoves dataset, predicting the Teacher TalkMoves labels. Our best configuration results in a macro-F1 of 0.764 and 0.756 Cohen's kappa, demonstrating improvements over direct prompting approaches and showing competitiveness with fine-tuned classifiers while gaining the benefit of greater transparency and comparability. To facilitate adoption and operationalization of the EduBehaviors framework in educational dialogue data by other researchers, we also release the EduBehaviors Toolkit with two complementary open-source tools based on the TalkMoves application.
Read the full abstract

All authors Julian Bernado, Ana Trindade Ribeiro, Xander Beberman, Susanna Loeb

Computation and Language

LexLattice: Multilingual Extractive Summarization via Neural Cellular Automata on Document Hierarchies

Sujay Uday Rittikar and Sheela Ramanna

Faithfulness is a central concern in legal text summarization, which motivates extractive approaches that select verbatim content traceable to its source. Such methods typically rank paragraphs or other structural units in isolation, yet give little attention to consolidating evidence that is distributed across, and shares salience between, distant parts of a document. We introduce LexLattice, an extractive summarizer that reifies a legal act's hierarchy as a two-dimensional semantic lattice and consolidates over it with a masked 2D neural cellular automata before selection. LexLattice attains state-of-the-art ROUGE across all 24 languages of EUR-Lex-Sum in both multilingual and cross-lingual settings, surpassing instruction-tuned baselines with billions of parameters, despite concentrating all trainable capacity in a 1.8M parameter consolidator over a frozen multilingual encoder. A consolidator trained only on high-resource languages further transfers to unseen languages with near-lossless retention (0.99), indicating that the model operates on language-agnostic semantic geometry rather than surface form. Our results position explicit consolidation over document structure as a compact and traceable alternative to scale for multilingual legal summarization.
Read the full abstract

Computation and Language · Cross-listed from cs.HC

ContraVis: Evidence-Grounded Visual Analytics for Contradiction Review in Legal Contracts

Luis Sante, Paula Lima, Mariana Rocha et al. (+1 more)

Legal contracts are structurally complex documents in which contradictions may emerge across distant and interconnected provisions. Although large language models (LLMs) improve legal language understanding, contradiction analysis remains a human-centered and evidence-grounded review task. We present ContraVis, a visual analytics system for human-in-the-loop contradiction analysis in legal contracts. The system models contracts as typed paragraph graphs that combine explicit contractual references with semantic relationships between paragraphs. This graph plays a dual role: it conditions LLM reasoning and serves as the interactive representation the analyst explores, keeping model context and human inspection aligned across coordinated views. In a controlled comparison, graph-conditioned reasoning recovered more injected contradictions than standalone LLM analysis as contract length grew, while surfacing additional candidates for analyst validation. A formative study with contract-domain lawyers indicated that in-context evidence comparison supported contradiction validation, and we distill design implications for evidence-grounded, LLM-assisted document review.
Read the full abstract

All authors Luis Sante, Paula Lima, Mariana Rocha, Jorge Poco

Computation and Language

LEGO: Synergizing Expert GraphRAG and Expert Chain-of-Thought for Legal Reasoning

Qingjing Chen, Junkai Zhang, Shaochun Wang et al. (+7 more)

Large language models are increasingly applied to high-risk domains such as law, yet complex legal reasoning remains limited by two structural challenges. First, existing RAG and GraphRAG methods emphasize lexical or semantic similarity while overlooking normative relations among legal provisions. Second, vanilla Chain-of-Thought prompting may generate plausible rationales without enforcing the normative structure of legal reasoning. To deal with the bottleneck of pipelines in the legal reasoning domain, we propose LEGO, a dual-module framework that synergizes Legal Expert GraphRAG and expert Chain-of-thought for complex legal reasoning. ExpertGraphRAG uses an expert-annotated civil code graph encoding these normative relations with a greedy normative-coverage retrieval algorithm to dynamically extract instance-specific provision subgraphs, while ExpertCoT organizes the retrieved provisions and case facts into structured Provision-Fact-Conclusion reasoning. With a Qwen3-8B backbone, LEGO achieves 40.53% exact-match accuracy on LawExamQA_Civil, outperforming the evaluated RAG and CoT baselines and performing comparably to the evaluated larger models, while remaining robust on multi-hop questions. It also achieves the best results among the evaluated baselines on the open-ended benchmarks. Ablation studies confirm the individual and complementary contributions of both modules, demonstrating LEGO's effectiveness in improving LLMs' complex legal reasoning ability. Code and dataset can be found in the link: https://github.com/BLK-WHT/LEGO
Read the full abstract

All authors Qingjing Chen, Junkai Zhang, Shaochun Wang, Jiahao Ding, Siyuan Zheng, Yukun Yan, Zhi Zheng, Antonino Rotolo, Yun Liu, Weixing Shen

Computation and Language

When Learned Context Planning Fails to Beat Strong Retrieval: A Controlled Study of Planning, Routing, and Reranking for Long-Context QA

Yingrui Li and Han Chen

Learned context planning selects evidence atoms before an answer model reasons over them. We test whether this learned selection improves long-context multiple-choice QA after strong retrieval, routing, budgeted-selector, and reranking controls. Our primary diagnostic uses all 503 LongBench-v2 MCQ questions with Qwen2.5-7B-Instruct. The planner is SFT-trained on outcome-selected traces from 140 training and 28 development questions; because the 503-question analysis includes those questions, it is partly transductive. At an 18k-character budget, anchored hybrid retrieval reaches 36.18% accuracy and BM25 reaches 35.98%, while the best direct planner-guided method reaches 34.19%. On the untouched 152-question test split, anchored hybrid remains higher (42.11% versus 36.84%). Leakage-safe routers cannot convert a large oracle gap. Under tight budgets, the best planner is ahead by only 0.40 points at 6k and loses at 9k; planner-guided reranking has a +1.79-point estimate at 6k with a paired interval crossing zero and ties the control at 9k. Packing-order and score-flatness analyses did not identify a stable mechanism. Under this setup, learned planning is a weak relevance signal rather than a replacement for strong retrieval.
Read the full abstract

Computation and Language

Classifying Interpretive Canons at the Sentence Level: A Benchmark from the German Federal Constitutional Court

Felix Ringe

Judicial reasoning remains challenging for large language models (LLMs) to analyze. This paper contributes a sentence-level benchmark for evaluating the ability of LLMs to classify interpretive canons as articulated by Larenz in the tradition of Savigny. Our contributions are threefold. First, we operationalize this conception of interpretation as classification criteria. Second, we provide a dataset of decisions of the German Federal Constitutional Court annotated at the sentence level. Third, we report baseline evaluations of four LLMs from three model families under expert hand-written prompts, compared against prompts optimized with Genetic-Pareto (GEPA). Mean F1 over the seven binary subtasks clusters between 70.4 and 79.2 across models, with grammatical interpretation usually the easiest canon to identify and systematic interpretation usually the hardest; under the tested configuration, GEPA-optimized prompts do not systematically outperform the hand-written ones, suggesting that the expert prompts provide a meaningful baseline.
Read the full abstract

Computation and Language

Recognized but Not Produced: A Generation Benchmark for Culturally Specific Kinship Terms

Sahil Pardasani and Madhusudan Singh

Current literature evaluates large language models (LLMs) on multilingual kinship understanding using multiple choice benchmarks, treating it as a recognition problem. We instead prompt five open weight LLMs to generate kinship terms in three non Western languages (Hindi, Tamil, and Korean) across two communicative tasks and pair this with a matched option-supported selection baseline. On identical relation language cells, GPT OSS120B selects the correct term in 90.67% of 75 valid cells but produces an accepted term in 36.00% of the corresponding attempts; Llama 3.370B shows the same pattern (77.92% versus 24.24%). Since the four-option condition displays the candidate terms and does not require script production, the difference is interpreted as an evaluation format gap rather than direct proof that lexical knowledge is intact. On explicitly specified L3 prompts, accuracy varies sharply, from GLM-5.1 at 72.29% to Llama-3.370B at 24.24%. The paternal-lineage advantage is language specific; it is large in Hindi but weak or reversed in Korean, while Tamil shared-term pairs provide a control for measurement variation. These results show that culturally specific kinship generation remains difficult even when the relationship is explicitly stated and motivate generation-based evaluation alongside multiple-choice testing.
Read the full abstract

Computation and Language · Cross-listed from cs.AI

Which Objectives Need a Dial? Predicting Objective Conflict and Covering Trade-offs in Steerable Pluralistic Alignment

David Tsoi and Esra Dönmez

People hold diverse, sometimes conflicting values, so no single aligned model can satisfy everyone. Pluralistic alignment therefore calls for steerable models that can balance competing objectives differently. Multi-Objective Direct Preference Optimization (MODPO) does this by using an objective weight to span a continuum of trade-offs. We study two questions: when can one model improve two objectives simultaneously, and how can many trade-offs be covered without training a separate model for each? Across seven objective pairs from HelpSteer and UltraFeedback, two pre-training measurements predict whether objectives align or conflict for human-annotated data, but not for AI-annotated data, where response length and repetition confound reward-model scores. For broader trade-off coverage, selecting the nearest trained model and merging model parameters both help, but neither consistently matches direct training. These findings yield practical guidance for building steerable models that serve diverse preferences.
Read the full abstract

Computation and Language

Experts Rise Where LLMs Disagree: Using Cross-Model Disagreement to Target Expert Effort in LLM Codebook Revision for Large-Scale Annotation

Zeyu He, Zhuqian Zhou, Kirk Vanacore et al. (+2 more)

Large-scale text annotation brings expert insight to millions of documents, often through a codebook that AI annotators follow. Developing a robust codebook, however, takes months. Large language models (LLMs) could speed this process by applying an early codebook to the data, surfacing cases with strong LLM disagreement, and eliciting expert feedback to address them. We examined three ways experts can provide feedback for LLM codebook revision: (i) editing LLM-generated revisions driven by cross-LLM disagreement (Codebook Verifying), (ii) answering questions about LLM disagreements (Question Answering), and (iii) labeling disagreement cases with rationales (Rationale Labeling). Experiments on thousands of tutoring-session transcripts show that Rationale Labeling yielded the highest LLM-labeling accuracy (64.9%) against expert labels, outperforming the expert-revised codebook (57.8%). The best Question Answering setting also outperformed it (60.5%). Our work shows that LLMs can be used to strategically target expert attention, shortening months of codebook revision to days without sacrificing labeling performance.
Read the full abstract

All authors Zeyu He, Zhuqian Zhou, Kirk Vanacore, Rene F. Kizilcec, Ting-Hao 'Kenneth' Huang

Computation and Language

COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference

Norah Alballa, Wenxuan Zhang, Salma Kharrat et al. (+4 more)

No single Large Language Model (LLM) is uniformly reliable across queries, motivating multi-model inference systems that either route among models or combine their outputs. However, routing stops after selecting an initial model, while dense collaboration invokes peers on every query. We show that collaboration is non-monotonic: peers can recover failures that no model solves alone, but can also corrupt initially correct answers. We introduce COMED (Controlled Model Escalation for Multi-LLM Deliberation), a post-anchor controller for selective cross-model collaboration. COMED uses anchor self-consistency, router margin, and a lightweight peer probe to accept confident answers, verify ambiguous cases, and escalate only when collaboration is likely beneficial. We formalize this trade-off with a rescue-harm decomposition showing that selective collaboration improves when rescued errors outweigh collaboration-induced harms. Across medical, scientific, and general reasoning benchmarks, COMED improves fixed and routed anchors in all 16 open-weight settings, with gains up to +10.7 percentage points on MedQA while invoking fewer models and using fewer decoded tokens than dense collaboration. On HLE with frontier models, COMED improves GPT-5.5 from 23.1% to 28.1%, outperforming dense collaboration and achieving the best results.
Read the full abstract

All authors Norah Alballa, Wenxuan Zhang, Salma Kharrat, Fares Fourati, Zafar Ayyub Qazi, Mohamed Elhoseiny, Marco Canini

Computation and Language · Cross-listed from cs.MM

Small Cues, Big Consequences: Learning Pivotal Cues for Multimodal Meme Classification

Akshit Sharma and Prashant W. Patil

Memes often derive their harmful, hateful, or sarcastic meaning from small but decisive visual, textual, or cross-modal cues. Existing multimodal classifiers can miss such evidence when relying mainly on global image-text representations. We introduce MemeCF, a cue-focused benchmark of 9,895 memes across harm, hate, and sarcasm, with annotations identifying the modality and rationale of the pivotal evidence. We also propose MemePIVOT, a local-global architecture for meme classification. MemePIVOT uses frozen CLIP features, unbalanced optimal transport to align words with image patches while allowing irrelevant evidence to remain unmatched, and an evidential fusion head to combine local grounding with global meme context under uncertainty. Experiments on HarMeme, PrideMM, and MemeCF show consistent gains over strong text-only, image-only, multimodal, and vision-language baselines. Cross-dataset and ablation results further show that explicit pivotal-evidence modeling improves robustness and contributes meaningfully beyond global multimodal representations. Our code and dataset are publicly available at https://github.com/AkshitSharma1/MemePIVOT
Read the full abstract

Computation and Language

Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs

Quan Nguyen-Tri, Mukul Ranjan and Zhiqiang Shen

Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive LLMs by enabling non-autoregressive text generation. However, their practical deployment remains limited by inefficient inference, largely due to the absence of effective Key-Value (KV) caching and scalable parallel decoding mechanisms. Existing acceleration methods typically study KV caching and parallel decoding in isolation, overlooking the I/O bottlenecks that arise when cache reuse and parallel token verification are jointly applied. In this work, we introduce $\textbf{Flash-dLLM}$, a training-free inference acceleration framework for fast and memory-efficient dLLMs. Flash-dLLM first identifies GPU memory I/O as a dominant bottleneck in KV-cache-enabled dLLM inference and addresses it with an I/O-aware fused KV-cache kernel that reduces redundant memory movement. Building on this optimized cache mechanism, Flash-dLLM further proposes an efficient KV-cache-driven draft-and-verify decoding strategy, where the dLLM itself serves as both drafter and verifier without requiring an auxiliary model. This unified design enables faster decoding while preserving generation quality and improving scalability to longer sequences and larger batch size. Extensive experiments on mathematical reasoning and code-generation benchmarks demonstrate that Flash-dLLM consistently outperforms existing state-of-the-art dLLM acceleration methods in both inference speed and memory efficiency. In particular, it achieves $5.1\times$ and $11.0\times$ speedups over prior strongest baseline Elastic-Cache on GSM8K and HumanEval, respectively.
Read the full abstract

Computation and Language

Agensh: Scaling Organizational Intelligence to 1,024 Agents

Zhihao Zhan, Ting Song, Li Dong et al. (+4 more)

A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent systems. However, the scalability of current multi-agent harnesses is often constrained by a central orchestrator's capacity to allocate tasks and coordinate workers. To address this limitation, we introduce Agensh, a scalable self-organized multi-agent harness without a central orchestrator: concurrent workers execute a multi-agent cooperation loop, continuously gathering context, claiming and self-assigning sub-tasks, taking action and sharing findings, verifying results, and merging progress in an asynchronous manner. The loop is supported by the agentic organization infrastructure comprising three components: a shared workspace holds proposed, ongoing, and completed work; a message interface lets workers communicate; and shared context retains reusable findings and work intentions. To test the scalability of Agensh, we evaluate it on the five hardest ProgramBench tasks with GPT-5.6-sol (high). Scaling from 1 to 128 agents raises the mean final test-pass rate from 19.31% to 28.78%, an approximately 49% relative improvement. Larger organizations reach comparable test-pass rates earlier. On pandoc, scaling from 1 to 1,024 agents raises the final test-pass rate from 33.89% to 55.06%. Worker trajectories further show that different forms of self-organized cooperation gradually emerges and standardizes as the organization grows. These results reveal the number of agents as a new scaling dimension for multi-agent organizations to expand the frontier of general intelligence, offering a practical solution for complex tasks under hard latency constraints or time budgets.
Read the full abstract

All authors Zhihao Zhan, Ting Song, Li Dong, Shaohan Huang, Jianxun Lian, Yan Xia, Furu Wei

Computation and Language

SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue

Haobo Zheng, Tan Tang, Yan Chen et al. (+2 more)

Long-term conversational memory in multi-party settings requires more than retrieving relevant content from long-term conversations: it must distinguish who said what, whom each statement concerns, how individuals perceive one another, what information is shared by the group, and how states change over time. Recent studies on multi-party dialogue benchmarks show that existing general-purpose LLM memory systems tend to lose person and group relations or struggle to integrate clues distributed across members, groups, and time. Together, these issues reveal two core bottlenecks: message attribution and relational understanding in multi-party dialogue, and state reconstruction from interleaved histories. To address both, we propose $\textbf{SpeakerMem-R1}$: its dual-track memory stores speaker-labeled verbatim messages and derived states organized into person-level and group-level views, then combines evidence from both tracks by entity, event, and time at query time. To reduce attribution and update errors during structured memory construction while enabling local deployment, we train Writer-R1 with SpeakerLevenshtein and speaker-conditioned GRPO. On GroupMemBench, SocialMemBench, and EverMemBench, SpeakerMem-R1 achieves binary accuracies of 47.9%, 69.2%, and 61.9%, respectively. On the publicly reported EverMemBench leaderboard from EverMind-AI, we achieves 62.33%, the best reported result among the latest state-of-the-art frameworks. It also achieves 70.85% on all 1,986 LoCoMo questions, which we use as a two-person long-term conversation boundary test. In a controlled evaluation of 305 questions, RL raises the SFT Writer's mean accuracy from 57.38% to 68.20%. We report both binary accuracy and token-F1, and ablations show that the verbatim and structured tracks, as well as person-level and group-level views, are complementary under the standardized evaluation interface.
Read the full abstract

All authors Haobo Zheng, Tan Tang, Yan Chen, Weijie Wang, Yingcai Wu

Computation and Language

Beyond Repeated Sampling: Learning Search Policies for LLM Reasoning

Ismail Labiad, Matthieu Kowalski, Marc Schoenauer et al. (+2 more)

Large language models increasingly tackle hard reasoning problems by spending more test-time compute, yet the dominant strategy remains naive repeated sampling: draw many independent solutions and hope one is correct. Because such sampling explores only through local decoding noise, it tends to produce many near duplicate attempts rather than genuinely different ideas. We ask whether exploration can instead be steered at a semantic level, by first sampling problem specific concepts, hints, or strategies and then conditioning answer generation on them. We refine this into a simple, more exploratory procedure that emits many diverse concepts in a single trajectory, and evaluate it on hard problems where repeated sampling struggles. We then go a step further and make concept generation trainable: a small concept generator is optimized with reinforcement learning so that its concepts maximize the downstream success of a larger, frozen answer generator. On hard mathematical reasoning problems, the trained concept generator substantially improves the answer generator's pass@k over naive repeated sampling at the same answer generation allocation, surpasses concepts drawn from much larger untuned models, and transfers to answer generators it was never trained against, including a model from a different family. A small model can thus be trained into an effective, reusable search policy for a much larger one.
Read the full abstract

All authors Ismail Labiad, Matthieu Kowalski, Marc Schoenauer, Rémi Munos, Julia Kempe

Computation and Language

Measuring the Serving Stack Instead of the Model: Hidden Confounds in Local Tool-Use Evaluation

Lijuan Tang and Yuemeng Zheng

A coding agent must emit a valid tool call–a parseable invocation of a tool in the provided schema–before the harness can execute its chosen action. We study how local serving stacks affect this protocol step and show that measured outcomes can depend on the serving layer rather than model behavior alone. In Ollama, the default tools= request is gated per model by a static template flag: some models are accepted and return calls as text, some return native tool_calls, while Phi-3 and Gemma-3 are rejected before inference. In our harness, rejection and retry exhaustion are not preserved as structured failure metadata, so downstream analysis can misclassify them as model non-calls and naively report 0% fidelity. Adding a text tool list while retaining the native channel recovers much of the measured fidelity for accepted models, whereas a uniform text protocol reduces fidelity for Llama-3.2, which has native tool-call support. Cross-stack probes on Ollama, llama.cpp, vLLM, and SGLang show different handling of the same request. Constrained decoding removes parse failures but can induce non-termination, and turn-pooled versus per-instance estimates differ by up to about 55 points. We conclude with a checklist for treating serving behavior as part of the evaluation protocol.
Read the full abstract

Computation and Language

Detecting GPT-Assisted Writing Using Interpretable Stylometric Features

Rajesh Kumar, Nabeel Siddiqui and Alexander Fuchsberger

Distinguishing GPT-assisted from independently authored student writing has become a critical challenge in academia. This paper evaluates the discriminative capability of interpretable stylometric features extracted solely from submitted text. Using data from 90 participants who wrote both independently and with ChatGPT assistance, we evaluate eight machine learning classifiers while keeping data from the same participant together during validation. On the held-out test set, Random Forest achieved an ROC-AUC of 0.87 and an F1-score of 0.84, with False Positive and False Negative rates of 22.2% and 11.1%, respectively. SHAP analysis shows that lexical and grammatical characteristics drive the resulting predictions. The findings suggest that transparent, text-intrinsic features provide measurable signal for detecting GPT-assisted writing.
Read the full abstract