Peer Revenue Impact factor: volatile

Papers

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

Computer Science Computation and Language

cs.CL Watch Computation and Language

Thursday 6 November

Computation and Language

Improving the Performance of Radiology Report De-identification with Large-Scale Training and Benchmarking Against Cloud Vendor Methods

Eva Prakash, Maayane Attias, Pierre Chambon et al. (+5 more)

Objective: To enhance automated de-identification of radiology reports by scaling transformer-based models through extensive training datasets and benchmarking performance against commercial cloud vendor systems for protected health information (PHI) detection. Materials and Methods: In this retrospective study, we built upon a state-of-the-art, transformer-based, PHI de-identification pipeline by fine-tuning on two large annotated radiology corpora from Stanford University, encompassing chest X-ray, chest CT, abdomen/pelvis CT, and brain MR reports and introducing an additional PHI category (AGE) into the architecture. Model performance was evaluated on test sets from Stanford and the University of Pennsylvania (Penn) for token-level PHI detection. We further assessed (1) the stability of synthetic PHI generation using a "hide-in-plain-sight" method and (2) performance against commercial systems. Precision, recall, and F1 scores were computed across all PHI categories. Results: Our model achieved overall F1 scores of 0.973 on the Penn dataset and 0.996 on the Stanford dataset, outperforming or maintaining the previous state-of-the-art model performance. Synthetic PHI evaluation showed consistent detectability (overall F1: 0.959 [0.958-0.960]) across 50 independently de-identified Penn datasets. Our model outperformed all vendor systems on synthetic Penn reports (overall F1: 0.960 vs. 0.632-0.754). Discussion: Large-scale, multimodal training improved cross-institutional generalization and robustness. Synthetic PHI generation preserved data utility while ensuring privacy. Conclusion: A transformer-based de-identification model trained on diverse radiology datasets outperforms prior academic and commercial systems in PHI detection and establishes a new benchmark for secure clinical text processing.
Read the full abstract

All authors Eva Prakash, Maayane Attias, Pierre Chambon, Justin Xu, Steven Truong, Jean-Benoit Delbrouck, Tessa Cook, Curtis Langlotz

Computation and Language

The truth is no diaper: Human and AI-generated associations to emotional words

Špela Vintar and Jan Jona Javoršek

Human word associations are a well-known method of gaining insight into the internal mental lexicon, but the responses spontaneously offered by human participants to word cues are not always predictable as they may be influenced by personal experience, emotions or individual cognitive styles. The ability to form associative links between seemingly unrelated concepts can be the driving mechanisms of creativity. We perform a comparison of the associative behaviour of humans compared to large language models. More specifically, we explore associations to emotionally loaded words and try to determine whether large language models generate associations in a similar way to humans. We find that the overlap between humans and LLMs is moderate, but also that the associations of LLMs tend to amplify the underlying emotional load of the stimulus, and that they tend to be more predictable and less creative than human ones.
Read the full abstract

Computation and Language

Plan of Knowledge: Retrieval-Augmented Large Language Models for Temporal Knowledge Graph Question Answering

Xinying Qian, Ying Zhang, Yu Zhao et al. (+3 more)

Temporal Knowledge Graph Question Answering (TKGQA) aims to answer time-sensitive questions by leveraging factual information from Temporal Knowledge Graphs (TKGs). While previous studies have employed pre-trained TKG embeddings or graph neural networks to inject temporal knowledge, they fail to fully understand the complex semantic information of time constraints. Recently, Large Language Models (LLMs) have shown remarkable progress, benefiting from their strong semantic understanding and reasoning generalization capabilities. However, their temporal reasoning ability remains limited. LLMs frequently suffer from hallucination and a lack of knowledge. To address these limitations, we propose the Plan of Knowledge framework with a contrastive temporal retriever, which is named PoK. Specifically, the proposed Plan of Knowledge module decomposes a complex temporal question into a sequence of sub-objectives from the pre-defined tools, serving as intermediate guidance for reasoning exploration. In parallel, we construct a Temporal Knowledge Store (TKS) with a contrastive retrieval framework, enabling the model to selectively retrieve semantically and temporally aligned facts from TKGs. By combining structured planning with temporal knowledge retrieval, PoK effectively enhances the interpretability and factual consistency of temporal reasoning. Extensive experiments on four benchmark TKGQA datasets demonstrate that PoK significantly improves the retrieval precision and reasoning accuracy of LLMs, surpassing the performance of the state-of-the-art TKGQA methods by 56.0% at most.
Read the full abstract

All authors Xinying Qian, Ying Zhang, Yu Zhao, Baohang Zhou, Xuhui Sui, Xiaojie Yuan

Computation and Language

T-FIX: Text-Based Explanations with Features Interpretable to eXperts

Shreya Havaldar, Weiqiu You, Chaehyeon Kim et al. (+12 more)

As LLMs are deployed in knowledge-intensive settings (e.g., surgery, astronomy, therapy), users are often domain experts who expect not just answers, but explanations that mirror professional reasoning. Yet evaluating whether an LLM "thinks like an expert" remains difficult: existing approaches rely on per-example expert annotation, making them costly, hard to scale, and tied to a single notion of correct reasoning within each domain. To address this gap, we introduce T-FIX, a unified evaluation framework that operationalizes expert alignment as a desired attribute of LLM-generated explanations. T-FIX spans seven scientific tasks across three domains, with each task evaluated against expert-defined criteria that capture domain-grounded reasoning rather than generic explanation quality. Our framework enables automatic, personalizable evaluation of expert alignment that generalizes to unseen explanations without ongoing expert involvement. Code is available at https://github.com/BrachioLab/FIX-2/.
Read the full abstract

All authors Shreya Havaldar, Weiqiu You, Chaehyeon Kim, Anton Xue, Helen Jin, Marco Gatti, Bhuvnesh Jain, Helen Qu, Amin Madani, Daniel A. Hashimoto, Gary E. Weissman, Rajat Deo, Sameed Khatana, Lyle Ungar, Eric Wong

Computation and Language · Cross-listed from cs.LG

DartQuant: Efficient Rotational Distribution Calibration for LLM Quantization

Yuantian Shao, Yuanteng Chen, Peisong Wang et al. (+5 more)

Quantization plays a crucial role in accelerating the inference of large-scale models, and rotational matrices have been shown to effectively improve quantization performance by smoothing outliers. However, end-to-end fine-tuning of rotational optimization algorithms incurs high computational costs and is prone to overfitting. To address this challenge, we propose an efficient distribution-aware rotational calibration method, DartQuant, which reduces the complexity of rotational optimization by constraining the distribution of the activations after rotation. This approach also effectively reduces reliance on task-specific losses, thereby mitigating the risk of overfitting. Additionally, we introduce the QR-Orth optimization scheme, which replaces expensive alternating optimization with a more efficient solution. In a variety of model quantization experiments, DartQuant demonstrates superior performance. Compared to existing methods, it achieves 47$\times$ acceleration and 10$\times$ memory savings for rotational optimization on a 70B model. Furthermore, it is the first to successfully complete rotational calibration for a 70B model on a single 3090 GPU, making quantization of large language models feasible in resource-constrained environments. Code is available at https://github.com/CAS-CLab/DartQuant.git.
Read the full abstract

All authors Yuantian Shao, Yuanteng Chen, Peisong Wang, Jianlin Yu, Jing Lin, Yiwu Yao, Zhihui Wei, Jian Cheng

Computation and Language · Cross-listed from cs.FL

Explorability in Pushdown Automata

Ayaan Bedi and Karoliina Lehtinen

We study explorability, a measure of nondeterminism in pushdown automata, which generalises history-determinism. An automaton is k-explorable if, while reading the input, it suffices to follow k concurrent runs, built step-by-step based only on the input seen so far, to construct an accepting one, if it exists. We show that the class of explorable PDAs lies strictly between history-deterministic and fully nondeterministic PDAs in terms of both expressiveness and succinctness. In fact increasing explorability induces an infinite hierarchy: each level k defines a strictly more expressive class than level k-1, yet the entire class remains less expressive than general nondeterministic PDAs. We then introduce a parameterized notion of explorability, where the number of runs may depend on input length, and show that exponential explorability precisely captures the context-free languages. Finally, we prove that explorable PDAs can be doubly exponentially more succinct than history-deterministic ones, and that the succinctness gap between deterministic and 2-explorable PDAs is not recursively enumerable. These results position explorability as a robust and operationally meaningful measure of nondeterminism for pushdown systems.
Read the full abstract

Computation and Language

WST: Weakly Supervised Transducer for Automatic Speech Recognition

Dongji Gao, Chenda Liao, Changliang Liu et al. (+5 more)

The Recurrent Neural Network-Transducer (RNN-T) is widely adopted in end-to-end (E2E) automatic speech recognition (ASR) tasks but depends heavily on large-scale, high-quality annotated data, which are often costly and difficult to obtain. To mitigate this reliance, we propose a Weakly Supervised Transducer (WST), which integrates a flexible training graph designed to robustly handle errors in the transcripts without requiring additional confidence estimation or auxiliary pre-trained models. Empirical evaluations on synthetic and industrial datasets reveal that WST effectively maintains performance even with transcription error rates of up to 70%, consistently outperforming existing Connectionist Temporal Classification (CTC)-based weakly supervised approaches, such as Bypass Temporal Classification (BTC) and Omni-Temporal Classification (OTC). These results demonstrate the practical utility and robustness of WST in realistic ASR settings. The implementation will be publicly available.
Read the full abstract

All authors Dongji Gao, Chenda Liao, Changliang Liu, Matthew Wiesner, Leibny Paola Garcia, Daniel Povey, Sanjeev Khudanpur, Jian Wu

Computation and Language

Abductive Inference in Retrieval-Augmented Language Models: Generating and Validating Missing Premises

Shiyin Lin

Large Language Models (LLMs) enhanced with retrieval – commonly referred to as Retrieval-Augmented Generation (RAG) – have demonstrated strong performance in knowledge-intensive tasks. However, RAG pipelines often fail when retrieved evidence is incomplete, leaving gaps in the reasoning process. In such cases, abductive inference – the process of generating plausible missing premises to explain observations – offers a principled approach to bridge these gaps. In this paper, we propose a framework that integrates abductive inference into retrieval-augmented LLMs. Our method detects insufficient evidence, generates candidate missing premises, and validates them through consistency and plausibility checks. Experimental results on abductive reasoning and multi-hop QA benchmarks show that our approach improves both answer accuracy and reasoning faithfulness. This work highlights abductive inference as a promising direction for enhancing the robustness and explainability of RAG systems.
Read the full abstract

Computation and Language · Cross-listed from cs.LG

Towards Scalable Meta-Learning of near-optimal Interpretable Models via Synthetic Model Generations

Kyaw Hpone Myint, Zhe Wu, Alexandre G.R. Day et al. (+1 more)

Decision trees are widely used in high-stakes fields like finance and healthcare due to their interpretability. This work introduces an efficient, scalable method for generating synthetic pre-training data to enable meta-learning of decision trees. Our approach samples near-optimal decision trees synthetically, creating large-scale, realistic datasets. Using the MetaTree transformer architecture, we demonstrate that this method achieves performance comparable to pre-training on real-world data or with computationally expensive optimal decision trees. This strategy significantly reduces computational costs, enhances data generation flexibility, and paves the way for scalable and efficient meta-learning of interpretable decision tree models.
Read the full abstract

All authors Kyaw Hpone Myint, Zhe Wu, Alexandre G.R. Day, Giri Iyengar

Computation and Language · Cross-listed from cs.AI

LLMs and Cultural Values: the Impact of Prompt Language and Explicit Cultural Framing

Bram Bulté and Ayla Rigouts Terryn

Large Language Models (LLMs) are rapidly being adopted by users across the globe, who interact with them in a diverse range of languages. At the same time, there are well-documented imbalances in the training data and optimisation objectives of this technology, raising doubts as to whether LLMs can represent the cultural diversity of their broad user base. In this study, we look at LLMs and cultural values and examine how prompt language and cultural framing influence model responses and their alignment with human values in different countries. We probe 10 LLMs with 63 items from the Hofstede Values Survey Module and World Values Survey, translated into 11 languages, and formulated as prompts with and without different explicit cultural perspectives. Our study confirms that both prompt language and cultural perspective produce variation in LLM outputs, but with an important caveat: While targeted prompting can, to a certain extent, steer LLM responses in the direction of the predominant values of the corresponding countries, it does not overcome the models' systematic bias toward the values associated with a restricted set of countries in our dataset: the Netherlands, Germany, the US, and Japan. All tested models, regardless of their origin, exhibit remarkably similar patterns: They produce fairly neutral responses on most topics, with selective progressive stances on issues such as social tolerance. Alignment with cultural values of human respondents is improved more with an explicit cultural perspective than with a targeted prompt language. Unexpectedly, combining both approaches is no more effective than cultural framing with an English prompt. These findings reveal that LLMs occupy an uncomfortable middle ground: They are responsive enough to changes in prompts to produce variation, but too firmly anchored to specific cultural defaults to adequately represent cultural diversity.
Read the full abstract

Computation and Language · Cross-listed from cs.MA

Multi-Agent Collaborative Framework For Math Problem Generation

Kia Karbasi, Kevin Hong, Mohammad Amin Samadi et al. (+1 more)

Automatic question generation (AQG) for mathematics education remains an elusive goal for Intelligent Tutoring Systems and educators. While pre-trained transformer-based language models have significantly advanced natural language generation, they often struggle to precisely control problem complexity and cognitive demands. In this paper, we introduce a collaborative multi-agent framework as a novel method of incorporating inference-time computation into AQG. This approach leverages multiple agents that iteratively refine generated question-answer pairs to better balance complexity and cognitive demand. We evaluate the generated questions on five meta-evaluation criteria: relevance, importance, clarity, difficulty matching, answerability, to assess the system's ability to control the required complexity and quality of the questions. Preliminary evaluations show that this collaborative multi-agent framework elevates the quality of generated educational content by fostering a more nuanced balance between cognitive challenge and clarity. These promising outcomes suggest that integrating collaborative multi-agent workflows can yield more controlled, pedagogically valuable content that can help advance automated educational content generation and adaptive learning environments.
Read the full abstract

All authors Kia Karbasi, Kevin Hong, Mohammad Amin Samadi, Gregory Pottie

Computation and Language

Direct Semantic Communication Between Large Language Models via Vector Translation

Fu-Chun Yang and Jason Eshraghian

In multi-agent settings, such as debate, reflection, or tool-calling, large language models (LLMs) pass messages as plain tokens, discarding most latent semantics. This constrains information transfer and adds unnecessary computational overhead. We form a latent bridge via vector translations, which use learned mappings that enable direct semantic exchange between representation spaces. A dual-encoder translator trained between Llama-2-7B and Mistral-7B-Instruct attains an average cosine alignment of 0.538. Injecting the translated vectors at 30 percent blending strength steers the target model's generation without destabilizing logits. Bidirectional evaluation shows a 2.01:1 transfer asymmetry, indicating that general-purpose models yield more transferable representations than instruction-tuned variants. This conservative injection preserves computational stability while demonstrating that cross-model latent communication is feasible, enabling collaborative AI systems that share meaning rather than tokens.
Read the full abstract

Computation and Language · Cross-listed from cs.SD

MIDI-LLM: Improving Text-to-MIDI Music Generation via Adapting Large Language Models

Shih-Lun Wu, Dave Carlton, Ryan Miyakawa et al. (+3 more)

We present MIDI-LLM, a recipe that improves multitrack text-to-MIDI generation via adapting Large Language Models (LLMs). MIDI-LLM expands an LLM's text vocabulary to include MIDI tokens and employs a two-stage training pipeline: (i) unimodal continued pretraining on music-adjacent text and standalone MIDIs, and (ii) multimodal supervised finetuning on text-MIDI pairs. Our instantiation of MIDI-LLM based on Llama 3.2 (1B) outperforms the recent Text2midi model in both text control and musical quality, and readily integrates with optimized inference ecosystems like vLLM. To align with real-world songwriting workflows, we further finetune our MIDI-LLM on the TheoryTab dataset for text-conditioned lead sheet (i.e., melody + chords) generation and infilling. A comprehensive ablation study validates the synergy between LLM text pretraining, standalone MIDI pretraining, and supervised text-to-MIDI finetuning. Finally, an in-the-wild blind user study conducted in a real-world creative workflow at scale with 58 Hookpad Aria users and 4,002 generated outputs demonstrates that our MIDI-LLM achieves the highest acceptance rate in zero-to-one lead sheet generation over baselines without text control or LLM pretraining, confirming its efficacy in human-AI music co-creation.
Read the full abstract

All authors Shih-Lun Wu, Dave Carlton, Ryan Miyakawa, Yoon Kim, Chris Donahue, Cheng-Zhi Anna Huang

Computation and Language · Cross-listed from cs.LG

RLHF: A comprehensive Survey for Cultural, Multimodal and Low Latency Alignment Methods

Raghav Sharma, Manan Mehta and Sai Tiger Raina

Reinforcement Learning from Human Feedback (RLHF) is the standard for aligning Large Language Models (LLMs), yet recent progress has moved beyond canonical text-based methods. This survey synthesizes the new frontier of alignment research by addressing critical gaps in multi-modal alignment, cultural fairness, and low-latency optimization. To systematically explore these domains, we first review foundational algo- rithms, including PPO, DPO, and GRPO, before presenting a detailed analysis of the latest innovations. By providing a comparative synthesis of these techniques and outlining open challenges, this work serves as an essential roadmap for researchers building more robust, efficient, and equitable AI systems.
Read the full abstract

Wednesday 5 November

Computation and Language

EvalCards: A Framework for Standardized Evaluation Reporting

Ruchira Dhar, Danae Sanchez Villegas, Antonia Karamolegkou et al. (+11 more)

Evaluation reports guide the use of general-purpose AI (GPAI) models, yet inconsistent reporting often leaves essential questions unanswered. Through a review of prior work, we identify three central issues in evaluation reporting practices today, reproducibility, accessibility, and accountability, and examine how existing reporting artifacts fall short in addressing them. We introduce the EvalCards framework, outlining its key dimensions and presenting a preliminary format for concise evaluation summaries. We conclude by discussing how the framework can guide the future development of evaluation reporting standards.
Read the full abstract

All authors Ruchira Dhar, Danae Sanchez Villegas, Antonia Karamolegkou, Alice Schiavone, Yifei Yuan, Xinyi Chen, Jiaang Li, Stella Frank, Laura De Grazia, Monorama Swain, Stephanie Brandl, Daniel Hershcovich, Anders Søgaard, Desmond Elliott

Computation and Language

COMPASS: Context-Modulated PID Attention Steering System for Hallucination Mitigation

Kenji Sahay, Snigdha Pandya, Rohan Nagale et al. (+4 more)

Large language models (LLMs) often generate fluent but factually incorrect statements despite having access to relevant evidence, a failure mode rooted in how they allocate attention between contextual and parametric knowledge. Understanding and steering this internal behavior is key both for trustworthy deployment and for scientific interpretability of model mechanisms. We introduce COMPASS (Context-Modulated PID Attention Steering System), a lightweight, interpretable control framework that embeds a model-based feedback loop directly within decoding. COMPASS quantifies context reliance via a transparent metric, the Context Reliance Score (CRS), which serves as an online probe of how attention heads ground generation in evidence. Using this interpretable signal, a PID controller dynamically modulates attention heads to maintain factual consistency without retraining or multi-pass decoding. Across benchmarks (HotpotQA, XSum, HaluEval, RAGTruth), COMPASS consistently reduces contextual hallucination rates (2.8 to 5.8 percent absolute) while revealing how distinct attention heads contribute to evidence alignment. These results highlight feedback-driven interpretability as a pathway toward scientific understanding of LLM behavior.
Read the full abstract

All authors Kenji Sahay, Snigdha Pandya, Rohan Nagale, Anna Lin, Shikhar Shiromani, Kevin Zhu, Dev Sunishchal

Computation and Language

Evaluating Modern Large Language Models on Low-Resource and Morphologically Rich Languages:A Cross-Lingual Benchmark Across Cantonese, Japanese, and Turkish

Chengxuan Xia, Qianye Wu, Hongbin Guan et al. (+3 more)

Large language models (LLMs) have achieved impressive results in high-resource languages like English, yet their effectiveness in low-resource and morphologically rich languages remains underexplored. In this paper, we present a comprehensive evaluation of seven cutting-edge LLMs – including GPT-4o, GPT-4, Claude~3.5~Sonnet, LLaMA~3.1, Mistral~Large~2, LLaMA-2~Chat~13B, and Mistral~7B~Instruct – on a new cross-lingual benchmark covering Cantonese, Japanese, and Turkish. Our benchmark spans four diverse tasks: open-domain question answering, document summarization, English-to-X translation, and culturally grounded dialogue. We combine human evaluations (rating fluency, factual accuracy, and cultural appropriateness) with automated metrics (e.g., BLEU, ROUGE) to assess model performance. Our results reveal that while the largest proprietary models (GPT-4o, GPT-4, Claude~3.5) generally lead across languages and tasks, significant gaps persist in culturally nuanced understanding and morphological generalization. Notably, GPT-4o demonstrates robust multilingual performance even on cross-lingual tasks, and Claude~3.5~Sonnet achieves competitive accuracy on knowledge and reasoning benchmarks. However, all models struggle to some extent with the unique linguistic challenges of each language, such as Turkish agglutinative morphology and Cantonese colloquialisms. Smaller open-source models (LLaMA-2~13B, Mistral~7B) lag substantially in fluency and accuracy, highlighting the resource disparity. We provide detailed quantitative results, qualitative error analysis, and discuss implications for developing more culturally aware and linguistically generalizable LLMs. Our benchmark and evaluation data are released to foster reproducibility and further research.
Read the full abstract

All authors Chengxuan Xia, Qianye Wu, Hongbin Guan, Sixuan Tian, Yilun Hao, Xiaoyu Wu

Computation and Language

A Super-Learner with Large Language Models for Medical Emergency Advising

Sergey K. Aityan, Abdolreza Mosaddegh, Rolando Herrero et al. (+12 more)

Medical decision-support and advising systems are critical for emergency physicians to quickly and accurately assess patients' conditions and make diagnosis. Artificial Intelligence (AI) has emerged as a transformative force in healthcare in recent years and Large Language Models (LLMs) have been employed in various fields of medical decision-support systems. We studied responses of a group of different LLMs to real cases in emergency medicine. The results of our study on five most renown LLMs showed significant differences in capabilities of Large Language Models for diagnostics acute diseases in medical emergencies with accuracy ranging between 58% and 65%. This accuracy significantly exceeds the reported accuracy of human doctors. We built a super-learner MEDAS (Medical Emergency Diagnostic Advising System) of five major LLMs - Gemini, Llama, Grok, GPT, and Claude). The super-learner produces higher diagnostic accuracy, 70%, even with a quite basic meta-learner. However, at least one of the integrated LLMs in the same super-learner produces 85% correct diagnoses. The super-learner integrates a cluster of LLMs using a meta-learner capable of learning different capabilities of each LLM to leverage diagnostic accuracy of the model by collective capabilities of all LLMs in the cluster. The results of our study showed that aggregated diagnostic accuracy provided by a meta-learning approach exceeds that of any individual LLM, suggesting that the super-learner can take advantage of the combined knowledge of the medical datasets used to train the group of LLMs.
Read the full abstract

All authors Sergey K. Aityan, Abdolreza Mosaddegh, Rolando Herrero, Haitham Tayyar, Jiang Han, Vikram Sawant, Qi Chen, Rishabh Jain, Aruna Senthamaraikannan, Stephen Wood, Manuel Mersini, Rita Lazzaro, Mario Balzanelli, Nicola Iacovazzo, Ciro Gargiulo Isacco

Computation and Language

Large Language Models for Scientific Idea Generation: A Creativity-Centered Survey

Fatemeh Shahhosseini, Arash Marioriyad, Ali Momen et al. (+3 more)

Scientific idea generation is central to discovery, requiring the joint satisfaction of novelty and scientific soundness. Unlike standard reasoning or general creative generation, scientific ideation is inherently open-ended and multi-objective, making its automation particularly challenging. Recent advances in large language models (LLMs) have enabled the generation of coherent and plausible scientific ideas, yet the nature and limits of their creative capabilities remain poorly understood. This survey provides a structured synthesis of methods for LLM-driven scientific ideation, focusing on how different approaches trade off novelty and scientific validity. We organize existing methods into five complementary families: External knowledge augmentation, Prompt-based distributional steering, Inference-time scaling, Multi-agent collaboration, and Parameter-level adaptation. To interpret their contributions, we adopt two complementary creativity frameworks: Boden taxonomy to characterize the expected level of creative novelty, and Rhodes 4Ps framework to analyze the aspects or sources of creativity emphasized by each method. By aligning methodological developments with cognitive creativity frameworks, this survey clarifies the evaluation landscape and identifies key challenges and directions for reliable and systematic LLM-based scientific discovery.
Read the full abstract

All authors Fatemeh Shahhosseini, Arash Marioriyad, Ali Momen, Mahdieh Soleymani Baghshah, Mohammad Hossein Rohban, Shaghayegh Haghjooy Javanmard

Computation and Language · Cross-listed from q-bio.NC

Approximating the Mathematical Structure of Psychodynamics

Bryce-Allen Bagley and Navin Khoshnan

The complexity of human cognition has meant that psychology makes more use of theory and conceptual models than perhaps any other biomedical field. To enable precise quantitative study of the full breadth of phenomena in psychological and psychiatric medicine as well as cognitive aspects of AI safety, there is a need for a mathematical formulation which is both mathematically precise and equally accessible to experts from numerous fields. In this paper we formalize human psychodynamics via the diagrammatic framework of process theory, describe its key properties, and explain the links between a diagrammatic representation and central concepts in analysis of cognitive processes in contexts such as psychotherapy, neurotechnology, AI alignment, AI agent representation of individuals in autonomous negotiations, developing human-like AI systems, and other aspects of AI safety.
Read the full abstract

Computation and Language

UTF-8 Plumbing: Byte-level Tokenizers Unavoidably Enable LLMs to Generate Ill-formed UTF-8

Preston Firestone, Shubham Ugare, Gagandeep Singh et al. (+1 more)

Subword tokenization segments input text according to a pre-defined vocabulary to feed it into a language model; the language model, in turn, generates a sequence made from this same vocabulary. The members of the vocabulary can be built of code points or bytes. Using code points means that all members of the vocabulary are valid UTF-8 characters. However, it also requires thousands of initial members to achieve acceptable coverage of inputs. Beginning with bytes, on the contrary, avoids out-of-vocabulary errors with only 256 initial members of the vocabulary, but the members of the vocabulary and sequences of them are not guaranteed to be valid UTF-8. Sequences that are not valid UTF-8 break code that assumes its input to be valid UTF-8. Applications of language models must account for the breakage thereby introduced. In this paper, we formalize tokenization using monoid theory and prove that tokenizers whose vocabularies contain tokens that are ill-formed UTF-8 can always produce sequences that are ill-formed UTF-8. We demonstrate formally that attempting to incrementally convert tokens back to a string and interpret the results as UTF-8 gives different results than converting the whole sequence of tokens at once. This formal result predicts real-world bugs: we evaluate mitigations for the problem identified and provide case studies of major foundation models, serving engines, and constrained generation systems.
Read the full abstract

All authors Preston Firestone, Shubham Ugare, Gagandeep Singh, Sasa Misailovic

Computation and Language

GEMMA-SQL: A Novel Text-to-SQL Model Based on Large Language Models

Hari Mohan Pandey, Anshul Gupta, Subham Sarkar et al. (+3 more)

Text-to-SQL systems enable users to interact with structured databases using natural language, eliminating the need for specialized programming knowledge. In this work, we introduce GEMMA-SQL, a lightweight and efficient text-to-SQL model built upon the open-source Gemma 2B architecture. Unlike many large language models (LLMs), GEMMA-SQL is fine-tuned in a resource-efficient, iterative manner and can be deployed on low-cost hardware. Leveraging the SPIDER benchmark for training and evaluation, GEMMA-SQL combines multiple prompting strategies, including few-shot learning, to enhance SQL query generation accuracy. The instruction-tuned variant, GEMMA-SQL Instruct, achieves 66.8% Test-Suite accuracy and 63.3% Exact Set Match accuracy, outperforming several state-of-the-art baselines such as IRNet, RYANSQL, and CodeXDavinci. The proposed approach demonstrates that effective prompt design and targeted instruction tuning can significantly boost performance while maintaining high scalability and adaptability. These results position GEMMA-SQL as a practical, open-source alternative for robust and accessible text-to-SQL systems.
Read the full abstract

All authors Hari Mohan Pandey, Anshul Gupta, Subham Sarkar, Minakshi Tomer, Schneider Johannes, Yan Gong

Computation and Language · Cross-listed from cs.CR

Jailbreaking in the Haystack

Rishi Rajesh Shah, Chen Henry Wu, Shashwat Saxena et al. (+3 more)

Recent advances in long-context language models (LMs) have enabled million-token inputs, expanding their capabilities across complex tasks like computer-use agents. Yet, the safety implications of these extended contexts remain unclear. To bridge this gap, we introduce NINJA (short for Needle-in-haystack jailbreak attack), a method that jailbreaks aligned LMs by appending benign, model-generated content to harmful user goals. Critical to our method is the observation that the position of harmful goals play an important role in safety. Experiments on standard safety benchmark, HarmBench, show that NINJA significantly increases attack success rates across state-of-the-art open and proprietary models, including LLaMA, Qwen, Mistral, and Gemini. Unlike prior jailbreaking methods, our approach is low-resource, transferable, and less detectable. Moreover, we show that NINJA is compute-optimal – under a fixed compute budget, increasing context length can outperform increasing the number of trials in best-of-N jailbreak. These findings reveal that even benign long contexts – when crafted with careful goal positioning – introduce fundamental vulnerabilities in modern LMs.
Read the full abstract

All authors Rishi Rajesh Shah, Chen Henry Wu, Shashwat Saxena, Ziqian Zhong, Alexander Robey, Aditi Raghunathan

Computation and Language

The Human Flourishing Geographic Index: A County-Level Dataset for the United States, 2013–2023

Stefano M. Iacus, Devika Jain, Andrea Nasuto et al. (+3 more)

Quantifying human flourishing, a multidimensional construct including happiness, health, purpose, virtue, relationships, and financial stability, is critical for understanding societal well-being beyond economic indicators. Existing measures often lack fine spatial and temporal resolution. Here we introduce the Human Flourishing Geographic Index (HFGI), derived from analyzing approximately 2.6 billion geolocated U.S. tweets (2013-2023) using fine-tuned large language models to classify expressions across 48 indicators aligned with Harvard's Global Flourishing Study framework plus attitudes towards migration and perception of corruption. The dataset offers monthly and yearly county- and state-level indicators of flourishing-related discourse, validated to confirm that the measures accurately represent the underlying constructs and show expected correlations with established indicators. This resource enables multidisciplinary analyses of well-being, inequality, and social change at unprecedented resolution, offering insights into the dynamics of human flourishing as reflected in social media discourse across the United States over the past decade.
Read the full abstract

All authors Stefano M. Iacus, Devika Jain, Andrea Nasuto, Giuseppe Porro, Marcello Carammia, Andrea Vezzulli

Computation and Language

Context informs pragmatic interpretation in vision-language models

Alvin Wei Ming Tan, Ben Prystawski, Veronica Boyce et al. (+1 more)

Iterated reference games - in which players repeatedly pick out novel referents using language - present a test case for agents' ability to perform context-sensitive pragmatic reasoning in multi-turn linguistic environments. We tested humans and vision-language models on trials from iterated reference games, varying the given context in terms of amount, order, and relevance. Without relevant context, models were above chance but substantially worse than humans. However, with relevant context, model performance increased dramatically over trials. Few-shot reference games with abstract referents remain a difficult task for machine learning models.
Read the full abstract

All authors Alvin Wei Ming Tan, Ben Prystawski, Veronica Boyce, Michael C. Frank

Computation and Language

GRAD: Graph-Retrieved Adaptive Decoding for Hallucination Mitigation

Manh Nguyen, Sunil Gupta, Dai Do et al. (+1 more)

Hallucination mitigation remains a persistent challenge for large language models (LLMs), even as model scales grow. Existing approaches often rely on external knowledge sources, such as structured databases or knowledge graphs, accessed through prompting or retrieval. However, prompt-based grounding is fragile and domain-sensitive, while symbolic knowledge integration incurs heavy retrieval and formatting costs. Motivated by knowledge graphs, we introduce Graph-Retrieved Adaptive Decoding (GRAD), a decoding-time method that grounds generation in corpus-derived evidence without retraining. GRAD constructs a sparse token transition graph by accumulating next-token logits across a small retrieved corpus in a single forward pass. During decoding, graph-retrieved logits are max-normalized and adaptively fused with model logits to favor high-evidence continuations while preserving fluency. Across three models and a range of question-answering benchmarks spanning intrinsic, extrinsic hallucination, and factuality tasks, GRAD consistently surpasses baselines, achieving up to 9.7$\%$ higher intrinsic accuracy, 8.6$\%$ lower hallucination rates, and 6.9$\%$ greater correctness compared to greedy decoding, while attaining the highest truth–informativeness product score among all methods. GRAD offers a lightweight, plug-and-play alternative to contrastive decoding and knowledge graph augmentation, demonstrating that statistical evidence from corpus-level token transitions can effectively steer generation toward more truthful and verifiable outputs.
Read the full abstract

All authors Manh Nguyen, Sunil Gupta, Dai Do, Hung Le

Computation and Language

Evaluating Machine Translation Datasets for Low-Web Data Languages: A Gendered Lens

Hellina Hailu Nigatu, Bethelhem Yemane Mamo, Bontu Fufa Balcha et al. (+5 more)

As low-resourced languages are increasingly incorporated into NLP research, there is an emphasis on collecting large-scale datasets. But in prioritizing quantity over quality, we risk 1) building language technologies that perform poorly for these languages and 2) producing harmful content that perpetuates societal biases. In this paper, we investigate the quality of Machine Translation (MT) datasets for three low-resourced languages–Afan Oromo, Amharic, and Tigrinya, with a focus on the gender representation in the datasets. Our findings demonstrate that while training data has a large representation of political and religious domain text, benchmark datasets are focused on news, health, and sports. We also found a large skew towards the male gender–in names of persons, the grammatical gender of verbs, and in stereotypical depictions in the datasets. Further, we found harmful and toxic depictions against women, which were more prominent for the language with the largest amount of data, underscoring that quantity does not guarantee quality. We hope that our work inspires further inquiry into the datasets collected for low-resourced languages and prompts early mitigation of harmful content. WARNING: This paper contains discussion of NSFW content that some may find disturbing.
Read the full abstract

All authors Hellina Hailu Nigatu, Bethelhem Yemane Mamo, Bontu Fufa Balcha, Debora Taye Tesfaye, Elbethel Daniel Zewdie, Ikram Behiru Nesiru, Jitu Ewnetu Hailu, Senait Mengesha Yayo

Computation and Language

Divide, Cache, Conquer: Dichotomic Prompting for Efficient Multi-Label LLM-Based Classification

Mikołaj Langner, Jan Eliasz, Ewa Rudnicka et al. (+1 more)

We introduce a method for efficient multi-label text classification with large language models (LLMs), built on reformulating classification tasks as sequences of dichotomic (yes/no) decisions. Instead of generating all labels in a single structured response, each target dimension is queried independently, which, combined with a prefix caching mechanism, yields substantial efficiency gains for short-text inference without loss of accuracy. To demonstrate the approach, we focus on affective text analysis, covering 24 dimensions including emotions and sentiment. Using LLM-to-SLM distillation, a powerful annotator model (DeepSeek-V3) provides multiple annotations per text, which are aggregated to fine-tune smaller models (HerBERT-Large, CLARIN-1B, PLLuM-8B, Gemma3-1B). The fine-tuned models show significant improvements over zero-shot baselines, particularly on the dimensions seen during training. Our findings suggest that decomposing multi-label classification into dichotomic queries, combined with distillation and cache-aware inference, offers a scalable and effective framework for LLM-based classification. While we validate the method on affective states, the approach is general and applicable across domains.
Read the full abstract

All authors Mikołaj Langner, Jan Eliasz, Ewa Rudnicka, Jan Kocoń

Computation and Language

STARS: Synchronous Token Alignment for Robust Supervision in Large Language Models

Mohammad Atif Quamar, Mohammad Areeb, Mikhail Kuznetsov et al. (+2 more)

Aligning large language models (LLMs) with human values is crucial for safe deployment. Inference-time techniques offer granular control over generation; however, they rely on model uncertainty, meaning an internal estimate of how likely the model believes its next tokens or outputs are correct, for segmentation. We show that this introduces two critical limitations: (a) vulnerability to miscalibrated confident hallucinations and (b) poor hardware utilization due to asynchronous, ragged batch processing. Together, these issues reduce alignment reliability while increasing token and compute costs, which limits their practical scalability. To address these limitations, building on dynamic inference-time alignment methods, we introduce STARS, Synchronous Token Alignment for Robust Supervision, a decoding-time algorithm, which steers generation by enforcing verification at fixed-horizon intervals. By decoupling segmentation from confidence, STARS enables lockstep parallel execution and robustly detects errors that uncertainty metrics miss. On the HH-RLHF benchmark, we demonstrate that STARS achieves competitive alignment quality with that of state-of-the-art dynamic methods, while strictly bounding rejection costs and maximizing system throughput. Furthermore, it outperforms fine-tuning and several state-of-the-art inference-time decoding strategies by good margins, and establishes fixed-horizon sampling as a robust, system-efficient alternative for aligning LLMs at scale. The code is publicly available at https://github.com/purseclab/STARS.
Read the full abstract

All authors Mohammad Atif Quamar, Mohammad Areeb, Mikhail Kuznetsov, Muslum Ozgur Ozmen, Z. Berkay Celik

Computation and Language · Cross-listed from cs.AI

How Different Tokenization Algorithms Impact LLMs and Transformer Models for Binary Code Analysis

Ahmed Mostafa, Raisul Arefin Nahid and Samuel Mulder

Tokenization is fundamental in assembly code analysis, impacting intrinsic characteristics like vocabulary size, semantic coverage, and extrinsic performance in downstream tasks. Despite its significance, tokenization in the context of assembly code remains an underexplored area. This study aims to address this gap by evaluating the intrinsic properties of Natural Language Processing (NLP) tokenization models and parameter choices, such as vocabulary size. We explore preprocessing customization options and pre-tokenization rules tailored to the unique characteristics of assembly code. Additionally, we assess their impact on downstream tasks like function signature prediction – a critical problem in binary code analysis. To this end, we conduct a thorough study on various tokenization models, systematically analyzing their efficiency in encoding assembly instructions and capturing semantic nuances. Through intrinsic evaluations, we compare tokenizers based on tokenization efficiency, vocabulary compression, and representational fidelity for assembly code. Using state-of-the-art pre-trained models such as the decoder-only Large Language Model (LLM) Llama 3.2, the encoder-only transformer BERT, and the encoder-decoder model BART, we evaluate the effectiveness of these tokenizers across multiple performance metrics. Preliminary findings indicate that tokenizer choice significantly influences downstream performance, with intrinsic metrics providing partial but incomplete predictability of extrinsic evaluation outcomes. These results reveal complex trade-offs between intrinsic tokenizer properties and their utility in practical assembly code tasks. Ultimately, this study provides valuable insights into optimizing tokenization models for low-level code analysis, contributing to the robustness and scalability of Natural Language Model (NLM)-based binary analysis workflows.
Read the full abstract

2511.03825 · 5 Nov · Publication Notice. This paper was published in the BAR 2025 Workshop (with NDSS 2025) and is for research and educational use. Copyright o̧pyright} 2025 Internet Society. All rights reserved. Personal/classroom reproduction is permitted with this notice and full paper citation. All other uses, including commercial, require prior written permission from the Internet Society

Abstract PDF $LANG ×3$TRFM ×2$ALGO ×1