AI Paper Atlas

Research map / Video / Video-language understanding

From Label Priors to Task Evidence: Long-Video Frame Selection via Bayesian GRPO

Shuochen Chang, Bingjie Gao, Qingyang Liu, Qianli Ma, Yibo Miao, Haonan Zhao, Zhaohe Liao, Xiaofeng Zhang, Jiangtong Li, Li Niu

NeurIPS 2026 · Video › Video-language understanding · approach: Long video understanding

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a Bayesian GRPO-based frame selector for long-video MLLMs that replaces static keyframe label priors with task-evidence signals from the downstream model.

Problem

Long-video understanding with MLLMs is limited by quadratic attention and context windows; existing retrieval/frame-selection methods rely on supervised learning over static keyframe annotations, which often do not match the information the downstream MLLM needs for reasoning.

Approach

Shifts frame selection from label-prior supervision to task-evidence optimization, training the selector with a Bayesian formulation and GRPO-style group-relative reinforcement learning so that selected frames are rewarded by their usefulness to the downstream MLLM.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace label reward with NLL task-evidence reward in Stage II (ablation)Bayesian GRPO selector · Video-MME (8 frames)accuracy: 58.3 -> 61.5Improved
add Stage II RL with label reward on top of label prior (ablation)Label Prior selector · Video-MME (32 frames)accuracy: 63.6 -> 63.8No clear effect
add Stage I label prior (pairwise ranking) to SigLIP2 selection (ablation)SigLIP2 selector · Video-MME (8 frames)accuracy: 54.7 -> 58.5Improved
add NLL-based Stage II on top of the label prior (ablation)Label Prior selector · MLVU avg (32 frames)accuracy: 68.2 -> 70.3Improved
use NLL dense reward instead of RLVR-style correctness reward (ablation)Bayesian GRPO selector · Video-MME (32 frames)accuracy: 62.1 -> 65.2Improved
use sparse correctness reward instead of label reward (ablation)Stage II reward · LongVideoBench (8 frames)accuracy: 57.6 -> 56.5Hurt
train Stage II on the target MLLM (InternVL-3-8B) instead of Qwen2.5-VL-7B (ablation)selector evaluated on InternVL-3-8B · Video-MME (32 frames)accuracy: 66.8 -> 66.5No clear effect
add NLL Stage II selector trained on frozen Qwen2.5-VL, applied to Qwen3-VL-8BQwen3-VL-8B with 32 frames · MLVU avgaccuracy: 65.3 -> 72.7Improved
increase frame budget K from 8 to 32 (ablation)Bayesian GRPO selector · Video-MME (Qwen2.5-VL-7B)accuracy: 61.5 -> 65.2Improved
increase frame budget K from 64 to 128 (ablation)Bayesian GRPO selector · MLVU avg (Qwen2.5-VL-7B)accuracy: 73.3 -> 73.4No clear effect
use the selector instead of uniform sampling with 8 framesQwen2.5-VL-7B · Video-MME overallaccuracy: 52.6 -> 61.5Improved
use the selector instead of uniform sampling at 32 framesInternVL-3.5-8B · Video-MME overallaccuracy: 61.1 -> 64.1Improved

Experimental setup

Builds on: GRPO

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy