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

Monday 15 December

Computation and Language

FIN-bench-v2: A Unified and Robust Benchmark Suite for Evaluating Finnish Large Language Models

Joona Kytöniemi, Jousia Piha, Akseli Reunamo et al. (+3 more)

We introduce FIN-bench-v2, a unified benchmark suite for evaluating large language models in Finnish. FIN-bench-v2 consolidates Finnish versions of widely used benchmarks together with an updated and expanded version of the original FIN-bench into a single, consistently formatted collection, covering multiple-choice and generative tasks across reading comprehension, commonsense reasoning, sentiment analysis, world knowledge, and alignment. All datasets are converted to HuggingFace Datasets, which include both cloze and multiple-choice prompt formulations with five variants per task, and we incorporate human annotation or review for machine-translated resources such as GoldenSwag and XED. To select robust tasks, we pretrain a set of 2.15B-parameter decoder-only models and use their learning curves to compute monotonicity, signal-to-noise, non-random performance, and model ordering consistency, retaining only tasks that satisfy all criteria. We further evaluate a set of larger instruction-tuned models to characterize performance across tasks and prompt formulations. All datasets, prompts, and evaluation configurations are publicly available via our fork of the Language Model Evaluation Harness at https://github.com/LumiOpen/lm-evaluation-harness. Supplementary resources are released in a separate repository at https://github.com/TurkuNLP/FIN-bench-v2.
Read the full abstract

All authors Joona Kytöniemi, Jousia Piha, Akseli Reunamo, Fedor Vitiugin, Farrokh Mehryary, Sampo Pyysalo

Computation and Language

MiniLingua: A Small Open-Source LLM for European Languages

Anna Aksenova, Boris Zverkov, Nicola Dainese et al. (+2 more)

Large language models are powerful but often limited by high computational cost, privacy concerns, and English-centric training. Recent progress demonstrates that small, efficient models with around one billion parameters can deliver strong results and enable on-device use. This paper introduces MiniLingua, a multilingual open-source LLM of one billion parameters trained from scratch for 13 European languages, designed to balance coverage and instruction-following capabilities. Based on evaluation results, the instruction-tuned version of MiniLingua outperforms EuroLLM, a model with a similar training approach but a larger training budget, on summarization, classification and both open- and closed-book question answering. Moreover, it remains competitive with more advanced state-of-the-art models on open-ended generation tasks. We release model weights, tokenizer and source code used for data processing and model training.
Read the full abstract

All authors Anna Aksenova, Boris Zverkov, Nicola Dainese, Alexander Nikitin, Pekka Marttinen

Computation and Language

Integrating Causal Reasoning into Automated Fact-Checking

Youssra Rebboud, Pasquale Lisena and Raphael Troncy

In fact-checking applications, a common reason to reject a claim is to detect the presence of erroneous cause-effect relationships between the events at play. However, current automated fact-checking methods lack dedicated causal-based reasoning, potentially missing a valuable opportunity for semantically rich explainability. To address this gap, we propose a methodology that combines event relation extraction, semantic similarity computation, and rule-based reasoning to detect logical inconsistencies between chains of events mentioned in a claim and in an evidence. Evaluated on two fact-checking datasets, this method establishes the first baseline for integrating fine-grained causal event relationships into fact-checking and enhance explainability of verdict prediction.
Read the full abstract

Computation and Language

AIR: Post-training Data Selection for Reasoning via Attention Head Influence

Jinrui Liu, Jeff Wu, Xuanguang Pan et al. (+3 more)

LLMs achieve remarkable multi-step reasoning capabilities, yet effectively transferring these skills via post-training distillation remains challenging. Existing data selection methods, ranging from manual curation to heuristics based on length, entropy, or overall loss, fail to capture the causal importance of individual reasoning steps, limiting distillation efficiency. To address this, we propose Attention Influence for Reasoning (AIR), a principled, unsupervised and training-free framework that leverages mechanistic insights of the retrieval head to select high-value post-training data. AIR first identifies reasoning-critical attention heads of an off-the-shelf model, then constructs a weakened reference model with disabled head influence, and finally quantifies the resulting loss divergence as the Attention Influence Score. This score enables fine-grained assessment at both the step and sample levels, supporting step-level weighted fine-tuning and global sample selection. Experiments across multiple reasoning benchmarks show that AIR consistently improves reasoning accuracy, surpassing heuristic baselines and effectively isolating the most critical steps and samples. Our work establishes a mechanism-driven, data-efficient approach for reasoning distillation in LLMs.
Read the full abstract

All authors Jinrui Liu, Jeff Wu, Xuanguang Pan, Gavin Cheung, Shuai Ma, Chongyang Tao

Computation and Language

AutoTool: Dynamic Tool Selection and Integration for Agentic Reasoning

Jiaru Zou, Ling Yang, Yunzhe Qi et al. (+5 more)

Agentic reinforcement learning has advanced large language models (LLMs) to reason through long chain-of-thought trajectories while interleaving external tool use. Existing approaches assume a fixed inventory of tools, which limits the adaptability of LLM agents to new or evolving toolsets. We present AutoTool, a training framework that equips LLM agents with dynamic tool-selection capabilities throughout their reasoning trajectories. AutoTool employs a dual-phase optimization pipeline: (i) SFT and RL-based trajectory stabilization for coherent reasoning, and (ii) KL-regularized Plackett-Luce Ranking to refine consistent multi-step tool selection. We further build a 200k dataset with explicit tool-selection rationales across 1,000+ tools and 100+ tasks spanning mathematics, science, code generation, and multimodal reasoning. Across ten diverse benchmarks, we train two base models, Qwen3-8B and Qwen2.5-VL-7B, with AutoTool. With fewer parameters, AutoTool consistently outperforms advanced LLM agents and tool-integration methods, yielding average gains of 6.4% in math & science reasoning, 4.5% in search-based QA, 7.7% in code generation, and 6.9% in multimodal understanding. In addition, AutoTool exhibits stronger generalization by dynamically leveraging unseen tools from evolving toolsets during inference.
Read the full abstract

All authors Jiaru Zou, Ling Yang, Yunzhe Qi, Sirui Chen, Mengting Ai, Ke Shen, Jingrui He, Mengdi Wang

Computation and Language

Efficient Adaptive Rejection Sampling for Accelerating Speculative Decoding in Large Language Models

Chendong Sun, Ali Mao, Lei Xu et al. (+1 more)

Speculative Decoding is a prominent technique for accelerating the autoregressive inference of large language models (LLMs) by employing a fast draft model to propose candidate token sequences and a large target model to verify them in parallel. However, its core component – the rejection sampling mechanism – relies on a fixed, context-independent random threshold. This leads to a significant "random rejection" problem in high-uncertainty generation scenarios, where plausible candidate tokens are frequently rejected due to random chance, undermining inference efficiency. This paper introduces Efficient Adaptive Rejection Sampling (EARS), a novel method that dynamically adjusts the acceptance threshold by incorporating the target model's own predictive uncertainty, measured as 1 - max(P_target). By introducing a tolerance term proportional to this uncertainty, EARS intelligently relaxes the acceptance criterion when the model is uncertain, effectively reducing random rejections while maintaining strict standards when the model is confident. Experiments on creative writing and open-domain QA tasks demonstrate that EARS significantly enhances the efficiency of speculative decoding, achieving up to an 18.12% increase in throughput with a negligible 0.84% accuracy drop on the GSM8K benchmark. The method requires no modifications to model architectures and can be seamlessly integrated into existing speculative decoding frameworks.
Read the full abstract

All authors Chendong Sun, Ali Mao, Lei Xu, mingmin Chen

Computation and Language · Cross-listed from cs.AI

SpeakRL: Synergizing Reasoning, Speaking, and Acting in Language Models with Reinforcement Learning

Emre Can Acikgoz, Jinoh Oh, Jie Hao et al. (+7 more)

Effective human-agent collaboration is increasingly prevalent in real-world applications. Current trends in such collaborations are predominantly unidirectional, with users providing instructions or posing questions to agents, where agents respond directly without seeking necessary clarifications or confirmations. However, the evolving capabilities of these agents require more proactive engagement, where agents should dynamically participate in conversations to clarify user intents, resolve ambiguities, and adapt to changing circumstances. Existing prior work under-utilize the conversational capabilities of language models (LMs), thereby optimizing agents as better followers rather than effective speakers. In this work, we introduce SpeakRL, a reinforcement learning (RL) method that enhances agents' conversational capabilities by rewarding proactive interactions with users, such as asking right clarification questions when necessary. To support this, we curate SpeakER, a synthetic dataset that includes diverse scenarios from task-oriented dialogues, where tasks are resolved through interactive clarification questions. We present a systematic analysis of reward design for conversational proactivity and propose a principled reward formulation for teaching agents to balance asking with acting. Empirical evaluations demonstrate that our approach achieves a 20.14% absolute improvement in task completion over base models without increasing conversation turns even surpassing even much larger proprietary models, demonstrating the promise of clarification-centric user-agent interactions.
Read the full abstract

All authors Emre Can Acikgoz, Jinoh Oh, Jie Hao, Joo Hyuk Jeon, Heng Ji, Dilek Hakkani-Tür, Gokhan Tur, Xiang Li, Chengyuan Ma, Xing Fan

Computation and Language · Cross-listed from cs.AI

MAC: A Multi-Agent Framework for Interactive User Clarification in Multi-turn Conversations

Emre Can Acikgoz, Jinoh Oh, Joo Hyuk Jeon et al. (+7 more)

Conversational agents often encounter ambiguous user requests, requiring an effective clarification to successfully complete tasks. While recent advancements in real-world applications favor multi-agent architectures to manage complex conversational scenarios efficiently, ambiguity resolution remains a critical and underexplored challenge–particularly due to the difficulty of determining which agent should initiate a clarification and how agents should coordinate their actions when faced with uncertain or incomplete user input. The fundamental questions of when to interrupt a user and how to formulate the optimal clarification query within the most optimal multi-agent settings remain open. In this paper, we propose MAC (Multi-Agent Clarification), an interactive multi-agent framework specifically optimized to resolve user ambiguities by strategically managing clarification dialogues. We first introduce a novel taxonomy categorizing user ambiguities to systematically guide clarification strategies. Then, we present MAC that autonomously coordinates multiple agents to interact synergistically with users. Empirical evaluations on MultiWOZ 2.4 demonstrate that enabling clarification at both levels increases task success rate 7.8% (54.5 to 62.3) and reduces the average number of dialogue turns (6.53 to 4.86) by eliciting all required user information up front and minimizing repetition. Our findings highlight the importance of active user interaction and role-aware clarification for more reliable human-agent communication.
Read the full abstract

All authors Emre Can Acikgoz, Jinoh Oh, Joo Hyuk Jeon, Jie Hao, Heng Ji, Dilek Hakkani-Tür, Gokhan Tur, Xiang Li, Chengyuan Ma, Xing Fan

Computation and Language

Uncovering the Role of Initial Saliency in U-Shaped Attention Bias: Scaling Initial Token Weight for Enhanced Long-Text Processing

Zewen Qiang, Sendong Zhao, Haochun Wang et al. (+2 more)

Large language models (LLMs) have demonstrated strong performance on a variety of natural language processing (NLP) tasks. However, they often struggle with long-text sequences due to the “lost in the middle” phenomenon. This issue has been shown to arise from a U-shaped attention bias, where attention is disproportionately focused on the beginning and end of a text, leaving the middle section underrepresented. While previous studies have attributed this bias to position encoding, our research first identifies an additional factor: initial saliency. It means that in the attention computation for each token, tokens with higher attention weights relative to the initial token tend to receive more attention in the prediction of the next token. We further find that utilizing this property by scaling attention weight between the initial token and others improves the model's ability to process long contexts, achieving a maximum improvement of 3.6% in MDQA dataset. Moreover, combining this approach with existing methods to reduce position encoding bias further enhances performance, achieving a maximum improvement of 3.4% in KV-Retrieval tasks.
Read the full abstract

All authors Zewen Qiang, Sendong Zhao, Haochun Wang, Bing Qin, Ting Liu

Computation and Language · Cross-listed from cs.CV

Heart Disease Prediction using Case Based Reasoning (CBR)

Mohaiminul Islam Bhuiyan, Chan Hue Wah, Nur Shazwani Kamarudin et al. (+2 more)

This study provides an overview of heart disease prediction using an intelligent system. Predicting disease accurately is crucial in the medical field, but traditional methods relying solely on a doctor's experience often lack precision. To address this limitation, intelligent systems are applied as an alternative to traditional approaches. While various intelligent system methods exist, this study focuses on three: Fuzzy Logic, Neural Networks, and Case-Based Reasoning (CBR). A comparison of these techniques in terms of accuracy was conducted, and ultimately, Case-Based Reasoning (CBR) was selected for heart disease prediction. In the prediction phase, the heart disease dataset underwent data pre-processing to clean the data and data splitting to separate it into training and testing sets. The chosen intelligent system was then employed to predict heart disease outcomes based on the processed data. The experiment concluded with Case-Based Reasoning (CBR) achieving a notable accuracy rate of 97.95% in predicting heart disease. The findings also revealed that the probability of heart disease was 57.76% for males and 42.24% for females. Further analysis from related studies suggests that factors such as smoking and alcohol consumption are significant contributors to heart disease, particularly among males.
Read the full abstract

All authors Mohaiminul Islam Bhuiyan, Chan Hue Wah, Nur Shazwani Kamarudin, Nur Hafieza Ismail, Ahmad Fakhri Ab Nasir

Computation and Language · Cross-listed from cs.AI

M-GRPO: Stabilizing Self-Supervised Reinforcement Learning for Large Language Models with Momentum-Anchored Policy Optimization

Bizhe Bai, Hongming Wu, Peng Ye et al. (+1 more)

Self-supervised reinforcement learning (RL) presents a promising approach for enhancing the reasoning capabilities of Large Language Models (LLMs) without reliance on expensive human-annotated data. However, we find that existing methods suffer from a critical failure mode under long-horizon training: a "policy collapse" where performance precipitously degrades. We diagnose this instability and demonstrate that simply scaling the number of rollouts – a common strategy to improve performance – only delays, but does not prevent, this collapse. To counteract this instability, we first introduce M-GRPO (Momentum-Anchored Group Relative Policy Optimization), a framework that leverages a slowly evolving momentum model to provide a stable training target. In addition, we identify that this process is often accompanied by a rapid collapse in policy entropy, resulting in a prematurely confident and suboptimal policy. To specifically address this issue, we propose a second contribution: an adaptive filtering method based on the interquartile range (IQR) that dynamically prunes low-entropy trajectories, preserving essential policy diversity. Our extensive experiments on multiple reasoning benchmarks demonstrate that M-GRPO stabilizes the training process while the IQR filter prevents premature convergence. The combination of these two innovations leads to superior training stability and state-of-the-art performance.
Read the full abstract

All authors Bizhe Bai, Hongming Wu, Peng Ye, Tao Chen

Computation and Language

LLM Rationalis? Measuring Bargaining Capabilities of AI Negotiators

Cheril Shah, Akshit Agarwal, Kanak Garg et al. (+1 more)

Bilateral negotiation is a complex, context-sensitive task in which human negotiators dynamically adjust anchors, pacing, and flexibility to exploit power asymmetries and informal cues. We introduce a unified mathematical framework for modeling concession dynamics based on a hyperbolic tangent curve, and propose two metrics burstiness tau and the Concession-Rigidity Index (CRI) to quantify the timing and rigidity of offer trajectories. We conduct a large-scale empirical comparison between human negotiators and four state-of-the-art large language models (LLMs) across natural-language and numeric-offers settings, with and without rich market context, as well as six controlled power-asymmetry scenarios. Our results reveal that, unlike humans who smoothly adapt to situations and infer the opponents position and strategies, LLMs systematically anchor at extremes of the possible agreement zone for negotiations and optimize for fixed points irrespective of leverage or context. Qualitative analysis further shows limited strategy diversity and occasional deceptive tactics used by LLMs. Moreover the ability of LLMs to negotiate does not improve with better models. These findings highlight fundamental limitations in current LLM negotiation capabilities and point to the need for models that better internalize opponent reasoning and context-dependent strategy.
Read the full abstract

All authors Cheril Shah, Akshit Agarwal, Kanak Garg, Mourad Heddaya

Computation and Language

An Open and Reproducible Deep Research Agent for Long-Form Question Answering

Ikuya Yamada, Wataru Ikeda, Ko Yoshida et al. (+5 more)

We present an open deep research system for long-form question answering, selected as a winning system in the text-to-text track of the MMU-RAG competition at NeurIPS 2025. The system combines an open-source large language model (LLM) with an open web search API to perform iterative retrieval, reasoning, and synthesis in real-world open-domain settings. To enhance reasoning quality, we apply preference tuning based on LLM-as-a-judge feedback that evaluates multiple aspects, including clarity, insightfulness, and factuality. Our experimental results show that the proposed method consistently improves answer quality across all three aspects. Our source code is publicly available at https://github.com/efficient-deep-research/efficient-deep-research.
Read the full abstract

All authors Ikuya Yamada, Wataru Ikeda, Ko Yoshida, Mengyu Ye, Hinata Sugimoto, Masatoshi Suzuki, Hisanori Ozaki, Jun Suzuki

Computation and Language · Cross-listed from cs.LG

Understanding Structured Financial Data with LLMs: A Case Study on Fraud Detection

Xuwei Tan, Yao Ma and Xueru Zhang

Detecting fraud in financial transactions typically relies on tabular models that demand heavy feature engineering to handle high-dimensional data and offer limited interpretability, making it difficult for humans to understand predictions. Large Language Models (LLMs), in contrast, can produce human-readable explanations and facilitate feature analysis, potentially reducing the manual workload of fraud analysts and informing system refinements. However, they perform poorly when applied directly to tabular fraud detection due to the difficulty of reasoning over many features, the extreme class imbalance, and the absence of contextual information. To bridge this gap, we introduce FinFRE-RAG, a two-stage approach that applies importance-guided feature reduction to serialize a compact subset of numeric/categorical attributes into natural language and performs retrieval-augmented in-context learning over label-aware, instance-level exemplars. Across four public fraud datasets and three families of open-weight LLMs, FinFRE-RAG substantially improves F1/MCC over direct prompting and is competitive with strong tabular baselines in several settings. Although these LLMs still lag behind specialized classifiers, they narrow the performance gap and provide interpretable rationales, highlighting their value as assistive tools in fraud analysis.
Read the full abstract

Computation and Language · Cross-listed from cs.LG

Scaling Bidirectional Spans and Span Violations in Attention Mechanism

Jongwook Kim, Sangheon Yun and Sukjin Yoon

The canonical $O(N^2)$ Transformer remains the empirical performance frontier in sequence modeling, and its training can be further optimized by addressing geometric inefficiency. We propose an optimization framework that leverages an asymmetric projection to decompose the backward-pass gradients into parallel spans and orthogonal violations, while keeping the canonical forward-pass $QKV$ structure intact. Through consistent experimental validation across various decomposition and projection setups, we provide strong theoretical evidence: the standard attention gradient is suboptimal. We demonstrated that selectively scaling these components, focusing primarily on $0^{th}$ order bidirectional parallel spans, yields the most effective learning signal. On the limited WikiText-2 dataset, and using a crude configuration, this method achieved a $0.56\%$ reduction in validation loss, confirming the framework's fundamental validity and suggesting significant potential gains on larger datasets and deeper training regimes
Read the full abstract

Computation and Language

Authors Should Label Their Own Documents

Marcus Ma, Cole Johnson, Nolan Bridges et al. (+3 more)

Third-party annotation is the status quo for labeling text, but egocentric information such as sentiment and belief can at best only be approximated by a third-person proxy. We introduce author labeling, an annotation technique where the writer of the document itself annotates the data at the moment of creation. We collaborate with a commercial chatbot with over 20,000 users to deploy an author labeling annotation system. This system identifies task-relevant queries, generates on-the-fly labeling questions, and records authors' answers in real time. We train and deploy an online-learning model architecture for product recommendation with author-labeled data to improve performance. We train our model to minimize the prediction error on questions generated for a set of predetermined subjective beliefs using author-labeled responses. Our model achieves a 537% improvement in click-through rate compared to an industry advertising baseline running concurrently. We then compare the quality and practicality of author labeling to three traditional annotation approaches for sentiment analysis and find author labeling to be higher quality, faster to acquire, and cheaper. These findings reinforce existing literature that annotations, especially for egocentric and subjective beliefs, are significantly higher quality when labeled by the author rather than a third party. To facilitate broader scientific adoption, we release an author labeling service for the research community at https://academic.echogroup.ai.
Read the full abstract

All authors Marcus Ma, Cole Johnson, Nolan Bridges, Jackson Trager, Georgios Chochlakis, Shrikanth Narayanan

Computation and Language

QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management

Weizhou Shen, Ziyi Yang, Chenliang Li et al. (+11 more)

We introduce QwenLong-L1.5, a model that achieves superior long-context reasoning capabilities through systematic post-training innovations. The key technical breakthroughs of QwenLong-L1.5 are as follows: (1) Long-Context Data Synthesis Pipeline: We develop a systematic synthesis framework that generates challenging reasoning tasks requiring multi-hop grounding over globally distributed evidence. By deconstructing documents into atomic facts and their underlying relationships, and then programmatically composing verifiable reasoning questions, our approach creates high-quality training data at scale, moving substantially beyond simple retrieval tasks to enable genuine long-range reasoning capabilities. (2) Stabilized Reinforcement Learning for Long-Context Training: To overcome the critical instability in long-context RL, we introduce task-balanced sampling with task-specific advantage estimation to mitigate reward bias, and propose Adaptive Entropy-Controlled Policy Optimization (AEPO) that dynamically regulates exploration-exploitation trade-offs. (3) Memory-Augmented Architecture for Ultra-Long Contexts: Recognizing that even extended context windows cannot accommodate arbitrarily long sequences, we develop a memory management framework with multi-stage fusion RL training that seamlessly integrates single-pass reasoning with iterative memory-based processing for tasks exceeding 4M tokens. Based on Qwen3-30B-A3B-Thinking, QwenLong-L1.5 achieves performance comparable to GPT-5 and Gemini-2.5-Pro on long-context reasoning benchmarks, surpassing its baseline by 9.90 points on average. On ultra-long tasks (1M~4M tokens), QwenLong-L1.5's memory-agent framework yields a 9.48-point gain over the agent baseline. Additionally, the acquired long-context reasoning ability translates to enhanced performance in general domains like scientific reasoning, memory tool using, and extended dialogue.
Read the full abstract

All authors Weizhou Shen, Ziyi Yang, Chenliang Li, Zhiyuan Lu, Miao Peng, Huashan Sun, Yingcheng Shi, Shengyi Liao, Shaopeng Lai, Bo Zhang, Dayiheng Liu, Fei Huang, Jingren Zhou, Ming Yan

Computation and Language

Building from Scratch: A Multi-Agent Framework with Human-in-the-Loop for Multilingual Legal Terminology Mapping

Lingyi Meng, Maolin Liu, Hao Wang et al. (+3 more)

Accurately mapping legal terminology across languages remains a significant challenge, especially for language pairs like Chinese and Japanese, which share a large number of homographs with different meanings. Existing resources and standardized tools for these languages are limited. To address this, we propose a human-AI collaborative approach for building a multilingual legal terminology database, based on a multi-agent framework. This approach integrates advanced large language models and legal domain experts throughout the entire process-from raw document preprocessing, article-level alignment, to terminology extraction, mapping, and quality assurance. Unlike a single automated pipeline, our approach places greater emphasis on how human experts participate in this multi-agent system. Humans and AI agents take on different roles: AI agents handle specific, repetitive tasks, such as OCR, text segmentation, semantic alignment, and initial terminology extraction, while human experts provide crucial oversight, review, and supervise the outputs with contextual knowledge and legal judgment. We tested the effectiveness of this framework using a trilingual parallel corpus comprising 35 key Chinese statutes, along with their English and Japanese translations. The experimental results show that this human-in-the-loop, multi-agent workflow not only improves the precision and consistency of multilingual legal terminology mapping but also offers greater scalability compared to traditional manual methods.
Read the full abstract

All authors Lingyi Meng, Maolin Liu, Hao Wang, Yilan Cheng, Qi Yang, Idlkaid Mohanmmed

Computation and Language · Cross-listed from physics.optics

Meta-GPT: Decoding the Metasurface Genome with Generative Artificial Intelligence

David Dang, Stuart Love, Meena Salib et al. (+7 more)

Advancing artificial intelligence for physical sciences requires representations that are both interpretable and compatible with the underlying laws of nature. We introduce METASTRINGS, a symbolic language for photonics that expresses nanostructures as textual sequences encoding materials, geometries, and lattice configurations. Analogous to molecular textual representations in chemistry, METASTRINGS provides a framework connecting human interpretability with computational design by capturing the structural hierarchy of photonic metasurfaces. Building on this representation, we develop Meta-GPT, a foundation transformer model trained on METASTRINGS and finetuned with physics-informed supervised, reinforcement, and chain-of-thought learning. Across various design tasks, the model achieves <3% mean-squared spectral error and maintains >98% syntactic validity, generating diverse metasurface prototypes whose experimentally measured optical responses match their target spectra. These results demonstrate that Meta-GPT can learn the compositional rules of light-matter interactions through METASTRINGS, laying a rigorous foundation for AI-driven photonics and representing an important step toward a metasurface genome project.
Read the full abstract

All authors David Dang, Stuart Love, Meena Salib, Quynh Dang, Samuel Rothfarb, Mysk Alnatour, Andrew Salij, Hou-Tong Chen, Ho Wai Lee, Wilton J.M. Kort-Kamp

Sunday 14 December

Computation and Language · Cross-listed from cs.IR

Comparative Analysis of Neural Retriever-Reranker Pipelines for Retrieval-Augmented Generation over Knowledge Graphs in E-commerce Applications

Teri Rumble, Zbyněk Gazdík, Javad Zarrin et al. (+1 more)

Recent advancements in Large Language Models (LLMs) have transformed Natural Language Processing (NLP), enabling complex information retrieval and generation tasks. Retrieval-Augmented Generation (RAG) has emerged as a key innovation, enhancing factual accuracy and contextual grounding by integrating external knowledge sources with generative models. Although RAG demonstrates strong performance on unstructured text, its application to structured knowledge graphs presents challenges: scaling retrieval across connected graphs and preserving contextual relationships during response generation. Cross-encoders refine retrieval precision, yet their integration with structured data remains underexplored. Addressing these challenges is crucial for developing domain-specific assistants that operate in production environments. This study presents the design and comparative evaluation of multiple Retriever-Reranker pipelines for knowledge graph natural language queries in e-Commerce contexts. Using the STaRK Semi-structured Knowledge Base (SKB), a production-scale e-Commerce dataset, we evaluate multiple RAG pipeline configurations optimized for language queries. Experimental results demonstrate substantial improvements over published benchmarks, achieving 20.4% higher Hit@1 and 14.5% higher Mean Reciprocal Rank (MRR). These findings establish a practical framework for integrating domain-specific SKBs into generative systems. Our contributions provide actionable insights for the deployment of production-ready RAG systems, with implications that extend beyond e-Commerce to other domains that require information retrieval from structured knowledge bases.
Read the full abstract

All authors Teri Rumble, Zbyněk Gazdík, Javad Zarrin, Jagdeep Ahluwalia

Computation and Language · Cross-listed from cs.LG

Social Determinants of Health Prediction for ICD-9 Code with Reasoning Models

Sharim Khan, Paul Landes, Adam Cross et al. (+1 more)

Social Determinants of Health correlate with patient outcomes but are rarely captured in structured data. Recent attention has been given to automatically extracting these markers from clinical text to supplement diagnostic systems with knowledge of patients' social circumstances. Large language models demonstrate strong performance in identifying Social Determinants of Health labels from sentences. However, prediction in large admissions or longitudinal notes is challenging given long distance dependencies. In this paper, we explore hospital admission multi-label Social Determinants of Health ICD-9 code classification on the MIMIC-III dataset using reasoning models and traditional large language models. We exploit existing ICD-9 codes for prediction on admissions, which achieved an 89% F1. Our contributions include our findings, missing SDoH codes in 139 admissions, and code to reproduce the results.
Read the full abstract

All authors Sharim Khan, Paul Landes, Adam Cross, Jimeng Sun

Computation and Language

DeliberationBench: When Do More Voices Hurt? A Controlled Study of Multi-LLM Deliberation Protocols

Vaarunay Kaushal and Taranveer Singh

Multi-agent systems where Large Language Models (LLMs) deliberate to form consensus have gained significant attention, yet their practical value over simpler methods remains under-scrutinized. We introduce DELIBERATIONBENCH, a controlled benchmark evaluating three deliberation protocols against a strong baseline of selecting the best response from a pool of model outputs. Across 270 questions and three independent seeds (810 total evaluations), we find a striking negative result: the best-single baseline achieves an 82.5% +- 3.3% win rate, dramatically outperforming the best deliberation protocol(13.8% +- 2.6%). This 6.0x performance gap is statistically significant (p < 0.01) and comes at 1.5-2.5x higher computational cost. Our findings challenge assumptions that complexity enhances quality in multi-LLM systems.
Read the full abstract

Computation and Language

AnimatedLLM: Explaining LLMs with Interactive Visualizations

Zdeněk Kasner and Ondřej Dušek

Large language models (LLMs) are becoming central to natural language processing education, yet materials showing their mechanics are sparse. We present AnimatedLLM, an interactive web application that provides step-by-step visualizations of a Transformer language model. AnimatedLLM runs entirely in the browser, using pre-computed traces of open LLMs applied on manually curated inputs. The application is available at https://animatedllm.github.io, both as a teaching aid and for self-educational purposes.
Read the full abstract

Computation and Language · Cross-listed from cs.CR

Auto-Tuning Safety Guardrails for Black-Box Large Language Models

Perry Abdulkadir

Large language models (LLMs) are increasingly deployed behind safety guardrails such as system prompts and content filters, especially in settings where product teams cannot modify model weights. In practice these guardrails are typically hand-tuned, brittle, and difficult to reproduce. This paper studies a simple but practical alternative: treat safety guardrail design itself as a hyperparameter optimization problem over a frozen base model. Concretely, I wrap Mistral-7B-Instruct with modular jailbreak and malware system prompts plus a ModernBERT-based harmfulness classifier, then evaluate candidate configurations on three public benchmarks covering malware generation, classic jailbreak prompts, and benign user queries. Each configuration is scored using malware and jailbreak attack success rate, benign harmful-response rate, and end-to-end latency. A 48-point grid search over prompt combinations and filter modes establishes a baseline. I then run a black-box Optuna study over the same space and show that it reliably rediscovers the best grid configurations while requiring an order of magnitude fewer evaluations and roughly 8x less wall-clock time. The results suggest that viewing safety guardrails as tunable hyperparameters is a feasible way to harden black-box LLM deployments under compute and time constraints.
Read the full abstract

Computation and Language · Cross-listed from cs.CV

SignRAG: A Retrieval-Augmented System for Scalable Zero-Shot Road Sign Recognition

Minghao Zhu, Zhihao Zhang, Anmol Sidhu et al. (+1 more)

Automated road sign recognition is a critical task for intelligent transportation systems, but traditional deep learning methods struggle with the sheer number of sign classes and the impracticality of creating exhaustive labeled datasets. This paper introduces a novel zero-shot recognition framework that adapts the Retrieval-Augmented Generation (RAG) paradigm to address this challenge. Our method first uses a Vision Language Model (VLM) to generate a textual description of a sign from an input image. This description is used to retrieve a small set of the most relevant sign candidates from a vector database of reference designs. Subsequently, a Large Language Model (LLM) reasons over the retrieved candidates to make a final, fine-grained recognition. We validate this approach on a comprehensive set of 303 regulatory signs from the Ohio MUTCD. Experimental results demonstrate the framework's effectiveness, achieving 95.58% accuracy on ideal reference images and 82.45% on challenging real-world road data. This work demonstrates the viability of RAG-based architectures for creating scalable and accurate systems for road sign recognition without task-specific training.
Read the full abstract

All authors Minghao Zhu, Zhihao Zhang, Anmol Sidhu, Keith Redmill

Computation and Language · Cross-listed from cs.CE

ERA-IT: Aligning Semantic Models with Revealed Economic Preference for Real-Time and Explainable Patent Valuation

Yongmin Yoo, Seungwoo Kim and Jingjiang Liu

Valuing intangible assets under uncertainty remains a critical challenge in the strategic management of technological innovation due to the information asymmetry inherent in high-dimensional technical specifications. Traditional bibliometric indicators, such as citation counts, fail to address this friction in a timely manner due to the systemic latency inherent in data accumulation. To bridge this gap, this study proposes the Economic Reasoning Alignment via Instruction Tuning (ERA-IT) framework. We theoretically conceptualize patent renewal history as a revealed economic preference and leverage it as an objective supervisory signal to align the generative reasoning of Large Language Models (LLMs) with market realities, a process we term Eco-Semantic Alignment. Using a randomly sampled dataset of 10,000 European Patent Office patents across diverse technological domains, we trained the model not only to predict value tiers but also to reverse-engineer the Economic Chain-of-Thought from unstructured text. Empirical results demonstrate that ERA-IT significantly outperforms both conventional econometric models and zero-shot LLMs in predictive accuracy. More importantly, by generating explicit, logically grounded rationales for valuation, the framework serves as a transparent cognitive scaffold for decision-makers, reducing the opacity of black-box AI in high-stakes intellectual property management.
Read the full abstract

Computation and Language

Counting Clues: A Lightweight Probabilistic Baseline Can Match an LLM

Furong Jia, Yuan Pu, Finn Guo et al. (+1 more)

Large language models (LLMs) excel on multiple-choice clinical diagnosis benchmarks, yet it is unclear how much of this performance reflects underlying probabilistic reasoning. We study this through questions from MedQA that ask for the most likely diagnosis. We introduce Frequency-Based Probabilistic Ranker (FBPR), a lightweight method that scores options with a smoothed Naive Bayes over concept-diagnosis co-occurrence statistics from a large corpus. When co-occurrence statistics are sourced from Dolma, the pretraining corpus for OLMo, and RedPajama, an open reproduction of the LLaMA pretraining-data recipe, FBPR achieves comparable performance to the corresponding LLMs. Furthermore, direct LLM inference and FBPR exhibit complementary strengths, correctly answering largely different subsets of questions with overlap barely above chance. Combining them with a simple hybrid method yields a notable gain over either method alone. These findings show that explicit probabilistic baselines can account for a substantial portion of the benchmark performance on diagnosis questions, while serving as historically grounded reference points and complementary signals for hybridization.
Read the full abstract

All authors Furong Jia, Yuan Pu, Finn Guo, Monica Agrawal

Computation and Language

What Matters in Evaluating Book-Length Stories? A Systematic Study of Long Story Evaluation

Dingyi Yang and Qin Jin

In this work, we conduct systematic research in a challenging area: the automatic evaluation of book-length stories (>100K tokens). Our study focuses on two key questions: (1) understanding which evaluation aspects matter most to readers, and (2) exploring effective methods for evaluating lengthy stories. We introduce the first large-scale benchmark, LongStoryEval, comprising 600 newly published books with an average length of 121K tokens (maximum 397K). Each book includes its average rating and multiple reader reviews, presented as critiques organized by evaluation aspects. By analyzing all user-mentioned aspects, we propose an evaluation criteria structure and conduct experiments to identify the most significant aspects among the 8 top-level criteria. For evaluation methods, we compare the effectiveness of three types: aggregation-based, incremental-updated, and summary-based evaluations. Our findings reveal that aggregation- and summary-based evaluations perform better, with the former excelling in detail assessment and the latter offering greater efficiency. Building on these insights, we further propose NovelCritique, an 8B model that leverages the efficient summary-based framework to review and score stories across specified aspects. NovelCritique outperforms commercial models like GPT-4o in aligning with human evaluations. Our datasets and codes are available at https://github.com/DingyiYang/LongStoryEval.
Read the full abstract

Computation and Language

Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects

Chris Latimer, Nicoló Boschi, Andrew Neeser et al. (+4 more)

Agent memory has been touted as a dimension of growth for LLM-based applications, enabling agents that can accumulate experience, adapt across sessions, and move beyond single-shot question answering. The current generation of agent memory systems treats memory as an external layer that extracts salient snippets from conversations, stores them in vector or graph-based stores, and retrieves top-k items into the prompt of an otherwise stateless model. While these systems improve personalization and context carry-over, they still blur the line between evidence and inference, struggle to organize information over long horizons, and offer limited support for agents that must explain their reasoning. We present Hindsight, a memory architecture that treats agent memory as a structured, first-class substrate for reasoning by organizing it into four logical networks that distinguish world facts, agent experiences, synthesized entity summaries, and evolving beliefs. This framework supports three core operations – retain, recall, and reflect – that govern how information is added, accessed, and updated. Under this abstraction, a temporal, entity aware memory layer incrementally turns conversational streams into a structured, queryable memory bank, while a reflection layer reasons over this bank to produce answers and to update information in a traceable way. On key long-horizon conversational memory benchmarks like LongMemEval and LoCoMo, Hindsight with an open-source 20B model lifts overall accuracy from 39% to 83.6% over a full-context baseline with the same backbone and outperforms full context GPT-4o. Scaling the backbone further pushes Hindsight to 91.4% on LongMemEval and up to 89.61% on LoCoMo (vs. 75.78% for the strongest prior open system), consistently outperforming existing memory architectures on multi-session and open-domain questions.
Read the full abstract

All authors Chris Latimer, Nicoló Boschi, Andrew Neeser, Chris Bartholomew, Gaurav Srivastava, Xuan Wang, Naren Ramakrishnan

Computation and Language

Does Tone Change the Answer? Evaluating Prompt Politeness Effects on Modern LLMs: GPT, Gemini, and LLaMA

Hanyu Cai, Binqi Shen, Lier Jin et al. (+2 more)

Prompt engineering has emerged as a critical factor influencing large language model (LLM) performance, yet the impact of pragmatic elements such as linguistic tone and politeness remains underexplored, particularly across different model families. In this work, we propose a systematic evaluation framework to examine how interaction tone affects model accuracy and apply it to three recently released and widely available LLMs: GPT-4o mini (OpenAI), Gemini 2.0 Flash (Google DeepMind), and Llama 4 Scout (Meta). Using the MMMLU benchmark, we evaluate model performance under Very Polite, Neutral, and Very Rude prompt variants across six tasks spanning STEM and Humanities domains, and analyze pairwise accuracy differences with statistical significance testing. Our results show that tone sensitivity is both model-dependent and domain-specific. Neutral or Very Polite prompts generally yield higher accuracy than Very Rude prompts, but statistically significant effects appear only in a subset of Humanities tasks, where rude tone reduces accuracy for GPT and Llama, while Gemini remains comparatively tone-insensitive. When performance is aggregated across tasks within each domain, tone effects diminish and largely lose statistical significance. Compared with earlier research, these findings suggest that dataset scale and coverage materially influence the detection of tone effects. Overall, our study indicates that while interaction tone can matter in specific interpretive settings, modern LLMs are broadly robust to tonal variation in typical mixed-domain use, providing practical guidance for prompt design and model selection in real-world deployments.
Read the full abstract

All authors Hanyu Cai, Binqi Shen, Lier Jin, Lan Hu, Xiaojing Fan