paper-with-me

Papers

Chains That See, Answers That Don't: A Multi-Aspect Evaluation Recipe for Forced Chain-of-Thought on Video-MME

2026-06-22 · Zhichao Fan, Yanhang Li, Zexin Zhuang arxiv

Forced chain-of-thought (CoT) is widely assumed to make vision-language models more reliable on video question answering. We propose a small three-probe evaluation recipe to test that assumption: paired accuracy across direct, CoT, answer-first, and no-video conditions; a counterfactual video-swap diagnostic over the CoT chains; and a four-rung visual-degradation ladder. Each probe is reported under both a strict and a permissive regex scorer, with multiplicity correction over a manuscript-declared primary family. Applied to Qwen2.5-VL on Video-MME subsets, the recipe returns a two-part finding. The CoT chains are strongly video-conditioned: swapping the input video collapses chain overlap and flips most final letters, the opposite of what a "boilerplate-chain" null would predict. Yet on the same data, forced CoT does not improve MCQ accuracy, and on the smaller 7B model it produces a small but statistically supported drop under a post-hoc primary scorer choice. We do not claim this generalizes beyond the Qwen2.5-VL / Video-MME instantiation; the raw responses and a single recomputation script will be released with the supplementary material so every number can be re-derived.

📄 PDF Abstract BibTeX arXiv:2606.22862

Code (0)

등록된 구현이 없습니다.

Tasks

Video Question Answering

Similar Papers 제목 키워드 기반

Composition Collapse: Stable Factual Knowledge Does Not Imply Compositional Reasoning

2026-05-26 · Zhe Yu, Wenpeng Xing, Yunzhao Wei, Jie Chen 외 arxiv

Post-training is routinely evaluated through aggregate benchmark scores that treat multi-hop reasoning as a single capability -- as if a model that answers more questions correctly must be better at assembling facts. We …

Latent Alignment of Procedural Concepts in Multimodal Recipes

2021-01-12 · WS 2020 7 · Hossein Rajaby Faghihi, Roshanak Mirzaee, Sudarshan Paliwal, Parisa Kordjamshidi

We propose a novel alignment mechanism to deal with procedural reasoning on a newly released multimodal QA dataset, named RecipeQA. Our model is solving the textual cloze task which is a reading comprehension on a recipe…

Question AnsweringReading Comprehension

Answering Questions by Meta-Reasoning over Multiple Chains of Thought

2023-04-25 · Ori Yoran, Tomer Wolfson, Ben Bogin, Uri Katz 외

Modern systems for multi-hop question answering (QA) typically break questions into a sequence of reasoning steps, termed chain-of-thought (CoT), before arriving at a final answer. Often, multiple chains are sampled and …

Multi-hop Question AnsweringQuestion Answering

Aggregation of Reasoning: A Hierarchical Framework for Enhancing Answer Selection in Large Language Models

2024-05-21 · Zhangyue Yin, Qiushi Sun, Qipeng Guo, Zhiyuan Zeng 외

Recent advancements in Chain-of-Thought prompting have facilitated significant breakthroughs for Large Language Models (LLMs) in complex reasoning tasks. Current research enhances the reasoning performance of LLMs by sam…

Answer Selection

A Critical Evaluation of Evaluations for Long-form Question Answering

2023-05-29 · Fangyuan Xu, Yixiao Song, Mohit Iyyer, Eunsol Choi

Long-form question answering (LFQA) enables answering a wide range of questions, but its flexibility poses enormous challenges for evaluation. We perform the first targeted study of the evaluation of long-form answers, c…

FormLong Form Question AnsweringQuestion AnsweringText Generation