AI Paper Atlas

Research map / Benchmarks / LVBench

LVBench leaderboard

56 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on LVBench. Best accuracy: 74.2 (Deep Video Discovery, NeurIPS 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

LVBench · accuracy

56 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
174.2Deep Video Discovery
OpenAI o3
Deep Video Discovery: Agentic Search with Tool Use for Long-form Video Understanding
NeurIPS 2025 · Table: Comparison on LVBench u
Video-language understanding
271.8Symphony
DeepSeek R1 + Seed 1.6 VL
Symphony: A Cognitively-Inspired Multi-Agent System for Long-Video Understanding
CVPR 2026 · Table 2
Video-language understanding
371.8MERIT
GPT-5
Keep It Simple: Multi-Key Episodic Memory Retrieval for Ultra-Long Video Understanding
ECCV 2026 · Table 2
Video-language understanding
468.6LensWalk (o3)
o3
LensWalk: Agentic Video Understanding by Planning How You See in Videos
CVPR 2026 · Table 1
Video-language understanding
561.9WorldMM-GPT
GPT-5
WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning
CVPR 2026 · Table 1
Video-language understanding
661.7TCoT
Gemini 1.5 Flash
Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames
NeurIPS 2025 · State-of-the-art comparison ta
Video-language understanding
761.5L2-VMAS
Qwen3-VL-8B-Thinking
Dual Latent Memory for Visual Multi-agent System
ICML 2026 · Table 4
LLM agents
861.5ReViSe
LLaVA-OV-7B
Towards Sparse Video Understanding and Reasoning
CVPR 2026 · Table 11
Video-language understanding
959.9CoTR (Ours)
Qwen3-VL-4B
Towards Temporal Compositional Reasoning in Long-Form Sports Videos
ECCV 2026 · Table 3
Video-language understanding
1058.4AdaCodec (comparable token budget)
Qwen3-VL-8B
AdaCodec: A Predictive Visual Code for Video MLLMs
NeurIPS 2026 · Table 1
Video-language understanding
46 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.