arXiv Computation and Language By Zheyu Huang, Zijing Shi, Haozhe Luo, Huadong Tang, Mingyu Liu, Meng Fang, Ling Chen

SocialReasonBench: A Video-QA Benchmark for Social Reasoning with Counterfactual Narrative Videos

Read the original on arXiv Computation and Language →

SocialReasonBench is a new video‑multiple‑choice QA benchmark designed to test socially grounded reasoning in interactive narrative videos. It uses branching gameplay footage from *Detroit: Become Human*, where player choices create alternative social outcomes that can be verified against the game’s script and flowchart. The benchmark includes seven reasoning dimensions—such as intent recognition, emotional empathy, moral dilemma, counterfactual reasoning, and causal antecedent—and employs a multi‑agent pipeline to curate clips, ground answer labels, and generate theory‑guided questions with diagnostic distractors.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI
Aug 11

Collaborative Multi-Agent Scripts Generation for Enhancing Imperfect-Information Reasoning in Murder Mystery Games

arXiv:2604. 11741v2 Announce Type: replace Abstract: Vision-language models (VLMs) have shown impressive capabilities in perceptual tasks, yet they degrade in complex multi-hop reasoning under multiplayer game settings with imperfect and deceptive information.

By Keyang Zhong, Junlin Xie, Hefeng Wu, Haofeng Li, Guanbin Li
arXiv Computation and Language
Aug 31

Cognitive Chain-of-Thought (CoCoT): Structured Multimodal Reasoning about Social Situations

The paper introduces Cognitive Chain-of-Thought (CoCoT), a structured reasoning framework for vision‑language models that divides multimodal social reasoning into three cognitively inspired stages: Perception, Situation, and Norm. CoCoT improves performance across diverse tasks—multimodal intent disambiguation, theory of mind, social commonsense reasoning, and safety instruction following—by 5.9% to 4.6% on average. Fine‑tuning on CoCoT‑structured traces further boosts accuracy by 5–6% without explicit prompting, indicating that models internalize the structured reasoning pattern and that the approach enhances interpretability and social alignment in multimodal systems.

By Eunkyu Park, Wesley Hanwen Deng, Gunhee Kim, Motahhare Eslami, Maarten Sap
arXiv AI
Sep 2

TempCloze: Can Video-LLMs Identify the Missing Middle?

TempCloze is a video cloze benchmark designed to evaluate visual temporal reasoning in Video-LLMs. The task presents a video’s beginning and ending clips and asks models to select the correct missing middle from four candidates, focusing on semantic, alignment, and progression aspects while minimizing appearance cues. Evaluation of 31 models shows that temporal alignment is the main challenge, with models performing better on semantic content and event progression but struggling to place events correctly in time.

By Wenqi Pei, Henry Hengyuan Zhao, Yilai Liu, Jiahao Meng, Han Chen, Ziyu Wang, Hongyang Du