AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

arXiv

MagicSelector: Joint Optimization for Agent Tool Selection via Counterfactual Decomposition and Progressive Reranking

HONOR Agentic Search Team, Zhengzong Chen, Lei Tang, Lijun Liu, Chuandi Jiang, Fan Yang, Keyun Chu, Chu Zhao, Shihao Liu, Minghang Li, Bo Liang, Can Wen, Hailong Wu, Jingnan Ju, Mian Liu, Nengbin Zhang, Peiqiang Wang, Penghe Nie, Qinhui Gu, Sijia Lv, Siqi Chen, Wei Zhang, Yang Xu, Yuhao Qian, Yuxiang Zhang, Zeng Cheng, Zhen Wang, Zuan Chen, Yuanyuan Zhao, Fei Huang

We present MagicSelector, a joint optimization framework integrating Counterfactual task decomposition, Progressive reranking, and Dynamic Top-K, designed to address the fundamental challenges of tool retrieval in agents. MagicSelector is a specialized framework capable of translating ambiguous user instructions into executable atomic subtasks and guiding high-precision tool retrieval, effectively mitigating redundant noise and severe context distraction in out-of-domain (OOD) scenarios. We empower MagicSelector with these capabilities through three key contributions: (1) a preference-guided counterfactual task decomposition mechanism that utilizes a counterfactual reward to quantify the marginal causal gain of decomposition on retrieval ranking, effectively imposing fine-grained structural supervision on logical coherence; (2) a progressive tool reranking method driven by self-distillation hard negative mining, which optimizes both point-wise and list-wise relevance to enhance fine-grained discrimination among highly similar tools; and (3) a dual semantic boundary-aware dynamic Top-K strategy that adaptively monitors reranking score cliffs and inter-tool semantic shifts to dynamically truncate the candidate list, maximizing relevant tool recall while filtering long-tail noise. Evaluated on MTDTool, the first task decomposition benchmark we constructed tailored for mobile multi-turn interactions with process-level annotations, MagicSelector yields promising performance. Extensive experiments demonstrate that MagicSelector significantly outperforms state-of-the-art methods in terms of tool retrieval accuracy, OOD generalization capability, and overall token efficiency, thereby demonstrating the effectiveness of our proposed framework.

arXiv

Measuring Human Value Expression in Social Media Texts: Calibrated LLM Annotation and Encoder Transfer

Maria Milkova, Maksim Rudnev

Measuring subjective constructs in naturally occurring social media text requires annotation procedures that are theoretically grounded, empirically validated, and transferable to an encoder model for scalable prediction. Using posts annotated according to Schwartz's theory of basic human values, we investigate how different LLMs and prompting strategies, which we call annotation regimes, operationalize the expression of values in text. Beyond standard classification metrics, we evaluate structural alignment, annotation ambiguity, error patterns, and stability across repeated runs. We find that different LLMs produce different value interpretations, and iterative prompt calibration through error analysis reduces misattributions and improves alignment with expert annotations. Error patterns are further used to derive targeted expert-verification rules for corpus annotation. We transfer soft LLM labels to an encoder model for prediction, retaining information about ambiguity in value expression. Finally, a sensitivity analysis on more than one million posts shows that regime-specific annotation differences propagate into predicted levels of value expression, whereas standardized temporal dynamics and the direction of major event responses are more robust.

arXiv

NorBERTo: A ModernBERT Model Trained for Portuguese with 331 Billion Tokens Corpus

Enzo S. N. Silva, Pablo B. Costa, Raphael C. Vlasman, Rosimeire P. Costa, Henrique L. P. Silva, Lucas F. A. O. Pellicer, Guilherme Rinaldo, Renato A. Almeida, Darian S. R. Rabbani, Cinthya O. Oestreich, Vinicius F. Caridá

High-quality corpora are essential for advancing Natural Language Processing (NLP) in Portuguese. Building on previous encoder-only models such as BERTimbau and Albertina PT-BR, we introduce NorBERTo, a modern encoder based on the ModernBERT architecture, featuring long-context support and efficient attention mechanisms. NorBERTo is trained on Aurora-PT, a newly curated Brazilian Portuguese corpus comprising 331 billion GPT-2 tokens collected from diverse web sources and existing multilingual datasets. We systematically benchmark NorBERTo against Strong baselines on semantic similarity, textual entailment and classification tasks using standardized datasets such as ASSIN 2 and PLUE. On PLUE, NorBERTo-large achieves the best results among the encoder models we evaluated, notably reaching 0.9191 F1 on MRPC and 0.7689 accuracy on RTE. On ASSIN 2, NorBERTo-large attains the highest entailment F1 (~0.904) among all encoders considered, although Albertina-900M and BERTimbau-large still hold an advantage. To the best of our knowledge, Aurora-PT is currently the largest openly available monolingual Portuguese corpus, surpassing previous resources. NorBERTo provides a modern, mid-sized encoder designed for realistic deployment scenarios: it is straight-forward to fine-tune, efficient to serve, and well suited as a backbone for retrieval-augmented generation and other downstream Portuguese NLP systems.

arXiv

HumorRank: A Tournament-Based Leaderboard for Evaluating Humor Generation in Large Language Models

Edward Ajayi, Prasenjit Mitra

Evaluating humor in large language models (LLMs) is an open challenge because existing approaches yield isolated, incomparable metrics rather than unified model rankings, making it difficult to track progress across systems. We introduce HumorRank, a tournament-based evaluation framework and leaderboard for textual humor generation. On two public benchmarks (SemEval-2026 MWAHAHA and Humor Transfer Bench), we conduct extensive automated pairwise evaluation across nine models spanning proprietary, open-weight, and specialized systems. Pairwise judgments are produced by LLM judges grounded in the General Theory of Verbal Humor (GTVH): each judge integrates structured comedic analysis into adjudication, jointly yielding a preference decision, an interpretable rationale, and mechanism, delivery, and failure tags rather than a black-box funniness score. Judgments are aggregated via an Adaptive Swiss tournament, with Bradley-Terry Maximum Likelihood Estimation (MLE) producing globally consistent humor generation capability rankings. Rankings are cross-judge stable: independent LLM judges (Llama 3.3 70B and Qwen 2.5 72B) yield Kendall tau = 0.889 on both benchmarks, and a human calibration study shows human-LLM agreement tracking human-human agreement on hard funny-versus-funny pairs. Our results demonstrate that HumorRank yields statistically grounded model stratifications, showing that humor quality is associated with mastery of comedic mechanisms such as incongruity, conciseness, escalation, and absurdity rather than model scale alone, with specialized fine-tuned models reaching parity with far larger systems. HumorRank thus provides a scalable, interpretable, and reproducible methodology for benchmarking and understanding LLM-generated humor.

arXiv

LLM2Vec-Gen: Generative Embeddings from Large Language Models

Parishad BehnamGhader, Vaibhav Adlakha, Fabian David Schmidt, Nicolas Chapados, Marius Mosbach, Siva Reddy

Fine-tuning LLM-based text embedders via contrastive learning maps inputs and outputs into a new representational space, discarding the LLM's output semantics. We propose LLM2Vec-Gen, a self-supervised alternative that instead produces embeddings directly in the LLM's output space by learning to represent the model's potential response. Specifically, trainable special tokens are appended to the input and optimized to compress the LLM's own response into a fixed-length embedding, guided by an unsupervised embedding teacher and a reconstruction objective. Crucially, the LLM backbone remains frozen and training requires only unlabeled queries. LLM2Vec-Gen achieves state-of-the-art self-supervised performance on the Massive Text Embedding Benchmark (MTEB), improving by 8.8% over the unsupervised embedding teacher. Since the embeddings preserve the LLM's response-space semantics, they inherit capabilities such as safety alignment (up to 22.6% reduction in harmful content retrieval) and reasoning (up to 35.6% improvement on reasoning-intensive retrieval). Finally, the learned embeddings are also interpretable: they can be decoded back into text to reveal their semantic content. Our code, data, and pretrained models are publicly available at https://github.com/McGill-NLP/llm2vec-gen.

arXiv

The Confidence Manifold: Geometric Structure of Correctness Representations in Language Models

Seonglae Cho, Zekun Wu, Kleyton Da Costa, Adriano Koshiyama

When a language model asserts that "the capital of Australia is Sydney," does it know this is wrong? Models assert misconceptions with the same fluency as facts, so the question cannot be answered from output uncertainty. Truth-related signals are known to exist in the residual stream, but not their geometry: how many dimensions carry the signal, how simple a detector can be, and whether it transfers. We characterize this geometry across 11 models (124M-14B) and test it causally with activation steering, concept erasure, and distributed alignment search. The structure is simple: two class centroids in a 2-8 dimensional subspace match a trained linear probe, and 25 labeled examples recover 90% of full-data AUC on GPT-2. Steering shifts hallucination rates by 9.1 points on six models, erasure drops detection to chance, and distributed alignment search, the only method that bounds rank, localizes at most five causal dimensions. The internal advantage is regime-specific: probes far outperform P(True) and semantic entropy on adversarial misconceptions but tie on standard QA. Single-dataset probes transfer near-randomly until joint multi-dataset training restores 0.73-0.91 AUC. That centroid distance matches probe performance indicates class separation is a mean shift, making detection geometric rather than learned.

arXiv

Scaling medical imaging report generation with multimodal reinforcement learning

Qianchu Liu, Sheng Zhang, Guanghui Qin, Yu Gu, Ying Jin, Sam Preston, Yanbo Xu, Sid Kiblawi, Wen-wai Yim, Timothy Ossowski, Tristan Naumann, Mu Wei, Hoifung Poon

Frontier models have demonstrated remarkable capabilities in understanding and reasoning with natural-language text, but they still exhibit major competency gaps in multimodal understanding and reasoning especially in high-value verticals such as biomedicine. Medical imaging report generation is a prominent example. Supervised fine-tuning can substantially improve performance, but they are prone to overfitting to superficial boilerplate patterns. In this paper, we introduce Universal Report Generation (UniRG) as a general framework for medical imaging report generation. By leveraging reinforcement learning as a unifying mechanism to directly optimize for evaluation metrics designed for end applications, UniRG can significantly improve upon supervised fine-tuning and attain durable generalization across diverse institutions and clinical practices. We trained UniRG-CXR on publicly available chest X-ray (CXR) data and conducted a thorough evaluation in CXR report generation with rigorous evaluation scenarios. On the authoritative ReXrank benchmark, UniRG-CXR sets new overall SOTA, outperforming prior state of the art by a wide margin. We release our model at https://huggingface.co/microsoft/UniRG-CXR.

arXiv

AfriEconQA: A Benchmark for Quantitative and Temporal Reasoning over World Bank Economic Reports

Edward Ajayi, Mustapha Alaba, David Stephen

Reliable question answering over long institutional documents requires more than topical retrieval: a system must localize the exact passage that supports a claim and preserve precise numerical and temporal detail when the same indicator recurs across years, countries, and projection horizons in dense, repetitive prose. Existing question-answering benchmarks rarely test this combination of exact grounding and temporal precision at document scale. We introduce AfriEconQA, a benchmark for document-grounded question answering built from 220 World Bank economic reports on African economies, a corpus whose claims are tied to specific fiscal periods, countries, and projection states. AfriEconQA contains 4,309 evidence-linked QA instances across five reasoning categories: Factoid (1,093), List (879), Multiple Choice (944), Synthesis (710), and Comparison (683), targeting quantitative extraction, set recovery, discrimination, causal integration, and cross-period or cross-country reasoning. Each instance carries sup- porting evidence and source provenance and is constructed through an agentic generation pipeline with evidence-grounding checks and a stratified human-validation subset for gold-label auditing. We evaluate Qwen 3.6 35B, DeepSeek v4-pro, and Gemma 4 12B IT under zero-shot, oracle, and hybrid retrieval-augmented generation (RAG) conditions on the held-out test split (n = 862). Across all three models, retrieval yields substantial gains, yet the best RAG system reaches only 0.545 F1, with residual errors concentrated in list extraction, synthesis, and temporally scoped comparison. AfriEconQA therefore poses a hard open challenge for exact numerical and temporal grounding over long institutional economic reports.

arXiv

Hallucinations and Truth: A Comprehensive Accuracy Evaluation of RAG, LoRA and DoRA

Mohammad Baqar, Rajat Khanda

Recent advancements in Generative AI have significantly improved the efficiency and adaptability of natural language processing (NLP) systems, particularly through Retrieval-Augmented Generation (RAG), Low-Rank Adaptation (LoRA), and Weight-Decomposed Low-Rank Adaptation (DoRA). RAG integrates external knowledge to enhance factual consistency in generative outputs, while LoRA enables parameter-efficient fine-tuning of large language models (LLMs). DoRA further refines this process by optimizing fine-tuning through adaptive parameter ranking and domain-aware weight adjustments, improving learning efficiency while maintaining inference performance. This paper presents a large-scale empirical evaluation of RAG, LoRA, and DoRA, with model fine-tuning and generation performance assessed on 20,000 FAQ-based queries, while the knowledge base spans 400,000 entries. The study analyzes key performance metrics such as accuracy, relevance, and inference latency. Experimental results demonstrate that DoRA achieves the highest accuracy (90.1%), relevance score (0.88), and lowest latency (110 ms per query), outperforming both LoRA and RAG in real-world, domain-specific generative AI applications. Furthermore, this study examines the trade-offs between fine-tuning efficiency, computational cost, and real-time adaptability across different models. Findings highlight RAG's effectiveness in knowledge grounding, LoRA's cost-efficient domain adaptation, and DoRA's ability to balance fine-tuning efficiency with model precision. These insights provide practical guidance for deploying AI-driven generative systems in accuracy-critical domains such as healthcare, finance, and legal services, ensuring scalability, reliability, and optimal performance in dynamic environments.

arXiv

Mental World Modeling

Hao Fei, Yiran Zhao

World models enable a predictive substrate for planning and action, yet existing formulations merely answer a physical question: what/where it is, and how will it evolve. Human behavior, however, is driven by hidden mental state (what a person believes, wants, intends, feels, and considers socially permissible), so a model that tracks the physical scene but not what each agent knows and believes about it predicts the wrong action for the right-looking scene. We formulate Mental World Modeling (MWM), a generic theoretical framework that makes mental variables core components of a world model rather than posthoc rationales: MWM aintains a coupled physical-mental world state, renders a target-specific partial observation, and simulates how candidate actions jointly update both components. We instantiate the framework in MENTIS, a training-free and fully inspectable baseline that decomposes the process into state parsing, target-observation generation, action decomposition, coupled physical and mental transition, and branch-level value evaluation. On a manually constructed, quality-controlled dataset of situated decision scenarios spanning text, image, and sounding-video stories, experiments with 8 modern LLM-based world models demonstrate that explicitly modeling the mental state is essential for predicting human decisions. Deeper analyses further expose the bottlenecks of current mental world modeling. We expect MWM as a next stage of world modeling, from simulating physical scenes to simulating the minds that act in them.

arXiv

Pangram 4 Technical Report

Ben Glickenhaus, Katherine Thai, Jenna Russell, Elyas Masrour, Yue Han, Max Spero, Bradley Emi

We present Pangram 4, the latest deep-learning-based AI-text classification model from Pangram Labs. We achieve an AUROC of 0.9916 with a false positive rate of 0.0041% and a false negative rate of 0.3396%. In addition to its increased overall accuracy compared with Pangram 3, Pangram 4 exhibits superior out-of-distribution generalization and robustness to adversarial attacks. Another novel contribution of Pangram 4 is its improved ability to distinguish fine-grained edits and mixed AI-human co-authored text. We demonstrate improvements to both boundary detection tasks and the detection of interleaved AI assistance. Finally, we report metrics on standard AI detection benchmarks showing that Pangram 4 achieves state-of-the-art performance on the AI text detection task across a wide variety of settings and domains.

arXiv

SpecFirst: Behavioral Specification Elicitation as a First-Class Step in Agent-Based Program Synthesis from Scratch

Yihao Chen, Shi Chang, Feng Lin, Khaled Chawa, Boyuan Chen, Shaowei Wang, Ahmed E. Hassan

LLM-based agents excel at software engineering tasks where an existing codebase provides context, but constructing a program from scratch remains fundamentally harder. Recent benchmarks such as ProgramBench quantify this gap: given only natural-language documentation and an execute-only binary as a behavioral oracle, even frontier models solve fewer than 1% of instances. Existing frameworks conflate documentation reading, behavioral exploration, and code synthesis into a single pass, causing agents to probe insufficiently, lose behavioral intent as context drifts, and propagate early misinterpretations into the final implementation. Inspired by classical requirements engineering, we argue that behavioral specification elicitation should be a first-class phase that precedes implementation. We present SpecFirst, a two-stage framework that forces the specification elicitation before code synthesis. A dedicated spec agent first probes the binary and combines observations with documentation into a structured specification. Next, a code synthesis agent then uses this specification to drive implementation. This decomposition resolves documentation ambiguities before coding begins and provides a stable behavioral reference throughout synthesis. We evaluate SpecFirst on all 200 ProgramBench instances across four models spanning two families and an order of magnitude of capability. SpecFirst consistently outperforms the single-loop baseline, improving test pass rates by 6.9%-21.3% and binary exploration coverage by 9.4%-18.5%, all statistically significant. Behavioral analysis on code synthesis further shows that a prior specification enables earlier and more sustained code construction. Our results demonstrate that an explicit requirements-engineering phase is an effective paradigm for from-scratch program construction.

arXiv

OmegaUse-OfficeVal: Benchmarking LLM Agents on Long-Horizon Office-Suite Tasks with Economic Grounding

Jingbo Zhou, Yusai Zhao, Qi Bao, Jingjia Cao, Zhenghai Chen, Chang Gao, Kaiqi Guo, Muxin Guo, Mingxuan Li, Xinjiang Lu, Yanru Ma, Yixiong Xiao, Zenghui Zhang, Le Zhang, Hua Wu

Large language model (LLM) agents are increasingly expected to assist users in completing tasks. However, existing benchmarks provide limited support for evaluating whether agents can carry out office-suite workflows at a reasonable cost. We introduce OmegaUse-OfficeVal, a benchmark for evaluating LLM agents on long-horizon office-suite tasks with task-level economic grounding. The benchmark comprises 100 tasks derived from office-suite requests proposed by practitioners and adapted through a privacy-preserving process. On average, these tasks require 2.32 hours of human labor to complete. An important feature of the benchmark is that each task is paired with two economic signals: human labor time and task price proxy. These signals enable direct comparisons between human costs and LLM inference costs, as well as value-weighted evaluation. To support stable evaluation, we develop code-based verifiers from fine-grained rubrics. We evaluate several frontier LLMs together with a human baseline. Although all evaluated LLMs are substantially cheaper and faster than human workers, they have not yet approached human-level deliverable quality. The code and dataset are fully open-sourced, and more information is available on our project website: https://omegause-officeval.github.io.

arXiv

MindForge: Teaching Small Language Models Whole-Life-Cycle Software Engineering via Source-Free Program Synthesis

Yihao Chen, Shi Chang, Khaled Chawa, Feng Lin, Boyuan Chen, Shaowei Wang, Ahmed E. Hassan

Coding agents have made substantial progress on software engineering tasks that modify existing codebases, including bug fixing and feature implementation. However, constructing a complete program from scratch remains a major challenge: even the frontier models evaluated on ProgramBench fully resolve fewer than 1% of tasks. One obstacle is the lack of scalable training environments for this from-scratch setting, spanning the whole software engineering life cycle, as existing environment-construction frameworks focus only on a single phase in software development. To address this gap, we introduce MindForge, an automated pipeline that converts open-source command-line programs into source-free environments that expose only a compiled reference executable and its documentation. Using MindForge, we construct training environments from repositories disjoint from those in ProgramBench, and curate a high-quality data recipe consisting of program synthesis trajectories using GLM-5.2 as the teacher agent. Fine-tuning Qwen3.6-27B on these trajectories increases its ProgramBench average test pass rate from 37.98% to 49.51%, achieving performance comparable to substantially larger frontier models. Moreover, the fine-tuned model consistently improves over the base model across all seven unseen software engineering benchmarks, spanning long-horizon repository generation and translation, bug fixing, feature implementation, and cross-language issue resolution, with absolute gains of 31.00 points on RepoZero-C2Rust, 14.16 on DeepSWE, 10.70/4.56 on NL2Repo-Bench (with/without tests), 5.04 on SWE-bench Verified, 5.93 on SWE-bench Pro, 5.22 on SWE-bench Multilingual, and 4.94 on FeatBench.

arXiv

Linguistic Monoculture in LLM-Assisted Language Use

Suhas Thejaswi, Juhi Kulshreshta, Lutz Oettershagen

Writing and communication are increasingly mediated by large language models (LLMs) that are being used to draft, revise and polish text. Although such assistance can improve clarity and help authors meet institutional expectations, widespread reliance on shared models may reduce population-level variation in linguistic form, a phenomenon we refer to as linguistic monoculture. We develop a mathematical framework in which authors and LLMs are represented as distributions over linguistic features and coevolve through repeated interaction. We analyze three interaction mechanisms: a shared model with a fixed linguistic distribution, a shared model recursively updated from author outputs, and personalized models updated through author-specific and population-level feedback. We characterize the resulting equilibria and convergence rates, showing that, shared models can drive authors toward a common norm, recursive feedback relocates the shared norm without altering pairwise spread under common conformity, and personalization can preserve a family of distinct author-model equilibria with nonzero linguistic diversity. We then endogenize conformity as a strategic choice trading off private benefits from clarity, legibility, and perceived fluency against distinctive style. Within this utility model, individually rational authors may conform more than is socially optimal because they do not internalize the value their distinctiveness provides to others, creating a negative externality and a price of monoculture that is finite for each fixed instance but can grow without bound when distinctiveness dominates authenticity. Synthetic simulations illustrate how fixed shared assistance, recursive feedback, and personalization produce different long-run diversity outcomes.

arXiv

On-Policy Distillation for LLM Safety: A Routing Approach to Template-Robust Realignment

Yongjian Guo, Wanlun Ma, Lingyu Shen, Xi Xiao, Sheng Wen

Fine-tuning is the dominant paradigm for specializing large language models (LLMs), yet it exposes a critical vulnerability: malicious data providers can embed harmful behaviors into downstream corpora, creating models that retain professional skills while violating human values on demand. Existing safety-realignment defenses often fail in practice due to three key limitations: they frequently cause catastrophic forgetting of specialized skills; their effectiveness collapses when the defender cannot observe the attacker's prompt template; and successfully realigned models remain susceptible to re-jailbreaking via simple system prompt switches. To address these challenges, we propose Routing-based On-Policy Distillation (ROPD), a novel realignment framework that models the divergence between aligned and compromised output probability distributions rather than fitting specific prompt templates. We conduct extensive experiments comparing ROPD against four state-of-the-art baselines across three datasets and three base models with varying alignment strengths. Our results demonstrate that when baseline defenses face template mismatches, often accompanied by severe degradation in downstream task performance. In contrast, ROPD substantially mitigates template-mismatch risks, maintaining superior robustness in both defense effectiveness and capability preservation. While our analysis indicates ROPD is not entirely immune to template shifts, its performance degradation is negligible compared to existing methods, establishing a new standard for robust LLM realignment.

arXiv

Evaluating Regional Bias in LLMs From Abstract Stereotype to Concrete Social Decision-Making

Jiayuan Di, Haoyi Yang, Yufei Luo, Jiahui Qu, Yiming Wang

Regional bias in large language models (LLMs) may shape both perceptions of regional groups and decisions about individuals from different regions. Yet existing studies often examine these manifestations separately, leaving their structure and consequences unclear. We introduce Stereotypes-to-Decisions (S2D), a systematic framework evaluating regional bias from abstract stereotypes to concrete social decisions. Covering all 34 provincial-level administrative regions of China, S2D evaluates six LLMs using stereotype ratings of Warmth (perceived friendliness and trustworthiness) and Competence (perceived capability and intelligence), along with paired-choice tasks across Education, Occupation, and Social Interaction. Results reveal substantial regional differences in regional scores, with considerable agreement across models, especially for Competence and Occupation decisions. Furthermore, these patterns are associated with regional economic and digital development indicators and display mixed human-like stereotypes, with some regions rated highly on one dimension but poorly on the other. They also remain largely stable across Chinese and English prompts. Overall, our findings show that regional bias in LLMs is prevalent, systematic, and consequential, motivating more regionally aware evaluation and mitigation.

arXiv

OptimismBench: Forecasting Bias and the Alignment Effect in Language Model Judgment

Seonglae Cho, Adriano Koshiyama

Large language models are increasingly used as decision aids whose probability judgments shape downstream choices. Whether those judgments carry a systematic directional tilt has been hard to detect: calibration metrics aggregate unsigned errors, and naturalistic uncertainty offers no ground-truth probability. When an LLM rates a startup's success at 70% but its failure at 15%, the missing 15 points expose a distortion no aggregate score flags. We introduce OptimismBench, which detects directional bias with inverted pairs: each scenario elicits both P(success) and P(failure), and asymmetry between the two framings yields a signed bias score without ground truth. Across 16 models from 8 providers, fourteen are optimistic; pessimism appears only in Anthropic's frontier tier. Eleven matched base-versus-chat pairs across four families show post-training sets the sign of the bias, with opposite shifts in different families. The pattern survives prompt, temperature, perspective, and self-debiasing ablations. A seventeen-model six-language comparison further shows model identity dominates language, with inter-model variance at 4.7x inter-language variance. We release 3,870 items across 10 languages for per-model directional-bias auditing. When alignment makes a model more helpful, it also tilts its probabilities; downstream pipelines inherit the tilt by default.

arXiv

Generation or Judgement? A Paradigm Perspective on LLM-Based Emotion-Cause Pair Extraction in Conversation

Weijie Feng, Hongchuang Wang, Binbin Liu, Zhiyong Cheng

Emotion-cause pair extraction in conversation (ECPEC) identifies utterance pairs in which one utterance causes an emotion expressed in another. Recent LLM-based approaches formulate ECPEC at markedly different granularities, ranging from generating complete pair sets to judging individual candidate pairs. In this paper, we make the surprising observation that task formulation substantially affects performance, where pair-level judgement outperforms dialogue-level generation in all 18 controlled comparisons. We investigate the sources of this paradigm gap and find that many relations omitted by dialogue-level generation remain recognizable under explicit pair queries, under which the model recognizes 92.7%-98.1% of emotion-cause relations. This suggests that LLMs can recognize emotion-cause relations but struggle to discover and return complete pair sets. Pair-level judgement alleviates this burden, although its candidate rankings are more reliable than the binary decisions produced by a shared threshold. Based on this diagnosis, we introduce an auxiliary retriever that selectively re-examines ambiguous boundary cases, yielding consistent F1 improvements of 0.50-1.46 points across three datasets while maintaining an inference time of only 1.49x that of the baseline paradigm. These findings show that task decomposition and candidate scope are critical to effectively utilizing LLMs for ECPEC.

arXiv

Credit Cards, Confusion, Computation, and Consequences: What Can We Uncover About Language Model Reasoning?

Arnav Hiray, Agam Shah, Caleb Lu, Meghaj Tarte, Harsit Mittal, Sudheer Chava

We introduce CreditCardQA, the first financial literacy benchmark for numerical reasoning derived from real credit card agreements. The dataset contains 1,800 questions, including first-person variants that reflect how consumers naturally ask about fees, interest, and payments. We evaluate a range of large language and reasoning models under Chain-of-Thought (CoT) and Program-of-Thought (PoT) prompting. Overall, PoT yields consistent performance gains, particularly for models with weaker baseline reasoning, and narrows gaps between open- and closed-source systems. Through error analysis, we show that failures arise less from arithmetic and more from misapplied financial rules, missed conditions, and misunderstandings of contractual terms. We further analyze question difficulty and find that comparisons, conditional logic, and monetary constraints are especially challenging. We also find that errors often arise in edge cases such as late-payment penalties or small-balance scenarios that are more likely to affect lower-income or financially vulnerable individuals.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.