LLM Digest

Story

arxiv_cs_ai · Nov 5, 2026 · paper

Source brief

AtariBench: Evaluating Spatiotemporal Reasoning of Vision-Language Models in Atari Gameplay

arxiv.orgNov 5, 2026
original source linked

In brief

Vision-Language Models (VLMs) have achieved strong performance on static visual understanding benchmarks, yet their ability to act in dynamic environments remains less well understood. We introduce AtariBench, a contr...

Feed lens

eval

Read the original at arxiv.org →Open in live feed Read that day’s brief

AtariBench: Evaluating Spatiotemporal Reasoning of Vision-Language Models in Atari Gameplay

Earlier in this thread 4 items

IWIBench: Evaluating Gradient-Free Model Driven Intra-weight Injection

Before You Continue: Evaluating Time-Sensitive Safety Awareness in Conversational Assistants

ProcessPrefBench: Evaluating Cross-Task Transfer of Individual Process Preferences from Naturalistic Human-AI Interaction

Heads That Write, Not Just Point: Image Retrieval Heads in Vision-Language Models