paper-with-me

Papers

Table-Text Alignment: Explaining Claim Verification Against Tables in Scientific Papers

2025-06-12 · Xanh Ho, Sunisth Kumar, Yun-Ang Wu, Florian Boudin, Atsuhiro Takasu, Akiko Aizawa

Scientific claim verification against tables typically requires predicting whether a claim is supported or refuted given a table. However, we argue that predicting the final label alone is insufficient: it reveals little about the model's reasoning and offers limited interpretability. To address this, we reframe table-text alignment as an explanation task, requiring models to identify the table cells essential for claim verification. We build a new dataset by extending the SciTab benchmark with human-annotated cell-level rationales. Annotators verify the claim label and highlight the minimal set of cells needed to support their decision. After the annotation process, we utilize the collected information and propose a taxonomy for handling ambiguous cases. Our experiments show that (i) incorporating table alignment information improves claim verification performance, and (ii) most LLMs, while often predicting correct labels, fail to recover human-aligned rationales, suggesting that their predictions do not stem from faithful reasoning.

📄 PDF Abstract BibTeX arXiv:2506.10486

Code (1)

alab-nii/scitabalign 공식 구현

Tasks

Claim Verification

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Enoki: Efficient Multi-Level Hallucination Detection

2026-09-01 · Elisei Rykov, Timur Ionov, Nikolay Ivanov, Maksim Savkin 외 arxiv

Ensuring factuality remains a critical challenge for deploying LLMs in high-stakes settings. Existing hallucination detectors usually operate at a single level: claim-level methods provide interpretable factual units, wh…

Open Information Extraction

AVeriTeC: A Dataset for Real-world Claim Verification with Evidence from the Web

2023-05-22 · NeurIPS 2023 11 · Michael Schlichtkrull, Zhijiang Guo, Andreas Vlachos

Existing datasets for automated fact-checking have substantial limitations, such as relying on artificial claims, lacking annotations for evidence and intermediate reasoning, or including evidence published after the cla…

Claim VerificationFact CheckingQuestion Answering

FactLens: Benchmarking Fine-Grained Fact Verification

2024-11-08 · Kushan Mitra, Dan Zhang, Sajjadur Rahman, Estevam Hruschka

Large Language Models (LLMs) have shown impressive capability in language generation and understanding, but their tendency to hallucinate and produce factually incorrect information remains a key limitation. To verify LL…

BenchmarkingFact VerificationText Generation

Distill and Align Decomposition for Enhanced Claim Verification

2026-02-25 · Jabez Magomere, Elena Kochkina, Samuel Mensah, Simerjot Kaur 외 arxiv

Complex claim verification requires decomposing sentences into verifiable subclaims, yet existing methods struggle to align decomposition quality with verification performance. We propose a reinforcement learning (RL) ap…

Reinforcement Learning

DeepSciVerify: Verifying Scientific Claim--Citation Alignment via LLM-Driven Evidence Escalation

2026-05-26 · Shaghayegh Sadeghi, Khashayar Khajavi, Rise Adhikari, Alexander Tessier arxiv

Misalignment between claims and their cited evidence is a common failure mode in reports generated by large language models, limiting their reliability in scientific and other high-stakes settings. We present DeepSciVeri…

Text Retrieval