English
Related papers

Related papers: FormInv: A Measurement Protocol for Semantic Invar…

200 papers

Ensembling Vision-Language Models (VLMs) from different providers maximizes benchmark accuracy, yet models from the same architectural family share correlated errors that standard voting ignores. We study this structure across 17 VLMs from…

Computer Vision and Pattern Recognition · Computer Science 2026-03-19 Zacharie Bugaud

Language models increasingly show their work by writing step-by-step reasoning before answering. But are these steps genuinely used, or is the answer rigid - fixed before reasoning begins? We introduce the Step-Level Reasoning Capacity…

Computation and Language · Computer Science 2026-04-14 Abhinaba Basu , Pavan Chakraborty

While Key-Value (KV) cache compression is essential for efficient LLM inference, current evaluations disproportionately focus on sparse retrieval tasks, potentially masking the degradation of High-Density Reasoning where Chain-of-Thought…

Computation and Language · Computer Science 2026-05-13 Xiang Liu , Zhenheng Tang , Hong Chen , Peijie Dong , Zeyu Li , Xiuze Zhou , Bo Li , Xuming Hu , Xiaowen Chu

Large language models can produce correct answers while relying on flawed reasoning traces, partly because common training objectives reward final-answer correctness rather than faithful intermediate reasoning. This undermines…

Artificial Intelligence · Computer Science 2026-01-06 Sanjeda Akter , Ibne Farabi Shihab , Anuj Sharma

Automatically filling Case Report Forms (CRFs) from clinical notes is challenging due to noisy language, strict output contracts, and the high cost of false positives. We describe our CL4Health 2026 submission for Dyspnea CRF filling (134…

Computation and Language · Computer Science 2026-04-23 Serhii Zabolotnii

Large Language Models for code generation frequently produce hallucinations in Fill-in-the-Middle (FIM) tasks -- plausible but incorrect completions such as invented API methods, invalid parameters, undefined variables, or non-existent…

Machine Learning · Computer Science 2026-05-12 Mahdi Erfanian , Nelson Daniel Troncoso , Aashna Garg , Amabel Gale , Xiaoyu Liu , Pareesa Ameneh Golnari , Shengyu Fu

LLMs show strong performance in code generation, but their outputs lack correctness guarantees. Sample-based uncertainty estimators address this by generating multiple candidate programs and measuring their disagreement. However, existing…

Software Engineering · Computer Science 2026-05-12 Weilin He , Arindam Sharma , Cristina David

In this paper, we describe SemEval-2013 Task 4: the definition, the data, the evaluation and the results. The task is to capture some of the meaning of English noun compounds via paraphrasing. Given a two-word noun compound, the…

Computation and Language · Computer Science 2019-11-26 Iris Hendrickx , Preslav Nakov , Stan Szpakowicz , Zornitsa Kozareva , Diarmuid Ó Séaghdha , Tony Veale

Multi-hop QA benchmarks frequently reward Large Language Models (LLMs) for spurious correctness, masking ungrounded or flawed reasoning steps. To shift toward rigorous reasoning, we propose SAFE, a dynamic benchmarking framework that…

Computation and Language · Computer Science 2026-04-03 Daeyong Kwon , Soyoung Yoon , Seung-won Hwang

Most Sign Language Translation (SLT) corpora pair each signed utterance with a single written-language reference, despite the highly non-isomorphic relationship between sign and spoken languages, where multiple translations can be equally…

Artificial Intelligence · Computer Science 2026-01-30 Václav Javorek , Tomáš Železný , Alessa Carbo , Marek Hrúz , Ivan Gruber

Evaluating AMR parsing accuracy involves comparing pairs of AMR graphs. The major evaluation metric, SMATCH (Cai and Knight, 2013), searches for one-to-one mappings between the nodes of two AMRs with a greedy hill-climbing algorithm, which…

Computation and Language · Computer Science 2019-05-31 Linfeng Song , Daniel Gildea

Many benchmarks for automated causal inference evaluate a system's performance based on a single numerical output, such as an Average Treatment Effect (ATE). This approach conflates two distinct steps in causal analysis: identification -…

Artificial Intelligence · Computer Science 2026-05-15 Ayush Sawarni , Jiyuan Tan , Vasilis Syrgkanis

Recent advances in reinforcement learning (RL) have led to substantial improvements in the mathematical reasoning abilities of LLMs, as measured by standard benchmarks. Yet these gains often persist even when models are trained with flawed…

Artificial Intelligence · Computer Science 2026-01-06 Jian Yao , Ran Cheng , Kay Chen Tan

Recent large language models (LLMs) achieve near-saturation accuracy on many established mathematical reasoning benchmarks, raising concerns about their ability to diagnose genuine reasoning competence. This saturation largely stems from…

Autoformalization aims to convert informal mathematical proofs into machine-verifiable formats, bridging the gap between natural and formal languages. However, ensuring semantic alignment between the informal and formalized statements…

Computation and Language · Computer Science 2024-10-15 Jianqiao Lu , Yingjia Wan , Yinya Huang , Jing Xiong , Zhengying Liu , Zhijiang Guo

The identification and localization of errors is a core task in peer review, yet the exponential growth of scientific output has made it increasingly difficult for human reviewers to reliably detect errors given the limited pool of experts.…

Computation and Language · Computer Science 2025-12-01 Sarina Xi , Vishisht Rao , Justin Payan , Nihar B. Shah

The miscalibration of Large Reasoning Models (LRMs) undermines their reliability in high-stakes domains, necessitating methods to accurately estimate the confidence of their long-form, multi-step outputs. To address this gap, we introduce…

Computation and Language · Computer Science 2026-01-22 Reza Khanmohammadi , Erfan Miahi , Simerjot Kaur , Ivan Brugere , Charese H. Smiley , Kundan Thind , Mohammad M. Ghassemi

Large Language Models (LLMs) have demonstrated remarkable reasoning abilities, yet existing test-time frameworks often rely on coarse self-verification and self-correction, limiting their effectiveness on complex tasks. In this paper, we…

Computation and Language · Computer Science 2025-11-14 Haizhou Shi , Ye Liu , Bo Pang , Zeyu Leo Liu , Hao Wang , Silvio Savarese , Caiming Xiong , Yingbo Zhou , Semih Yavuz

Small changes to how a buyer phrases a question -- "best CRM" vs "top CRM" vs "best CRM for a SaaS startup" -- produce substantially different brand recommendations from AI assistants. Across ~6,000 paraphrase runs and ~6,000 same-prompt…

Information Retrieval · Computer Science 2026-05-28 Will Jack , Noah Lehman , Keller Maloney , Sarah Xu

Benchmark scores for Large Language Models (LLMs) can be inflated by memorization of test items or near duplicates. We present a simple, protocol that probes generalization by re-evaluating models on paraphrased versions of benchmark…

Computation and Language · Computer Science 2025-10-13 Juan Miguel Navarro Carranza