# TimeSearch-R: Adaptive Temporal Search for Long-Form Video Understanding via Self-Verification Reinforcement Learning

arXiv: https://arxiv.org/abs/2511.05489 · HTML: https://mlsanity.com/paper/2511.05489/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 32 of its 43 references that are in the map.
- First posted on arXiv 7 Nov 2025.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Builds on (most cited first, 30 of 32)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning](https://mlsanity.com/paper/2503.09516/) (2025)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [YOLO-World: Real-Time Open-Vocabulary Object Detection](https://mlsanity.com/paper/2401.17270/) (2024)
- [Video-R1: Reinforcing Video Reasoning in MLLMs](https://mlsanity.com/paper/2503.21776/) (2025)
- [DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning](https://mlsanity.com/paper/2505.14362/) (2025)
- [VideoAgent: Long-form Video Understanding with Large Language Model as Agent](https://mlsanity.com/paper/2403.10517/) (2024)
- [Visual Sketchpad: Sketching as a Visual Chain of Thought for Multimodal Language Models](https://mlsanity.com/paper/2406.09403/) (2024)
- [Counterfactual VQA: A Cause-Effect Look at Language Bias](https://mlsanity.com/paper/2006.04315/) (2020)
- [VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos](https://mlsanity.com/paper/2405.19209/) (2024)
- [LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL](https://mlsanity.com/paper/2503.07536/) (2025)
- [Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontiers](https://mlsanity.com/paper/2506.23918/) (2025)
- [CinePile: A Long Video Question Answering Dataset and Benchmark](https://mlsanity.com/paper/2405.08813/) (2024)
- [Frame-Voyager: Learning to Query Frames for Video Large Language Models](https://mlsanity.com/paper/2410.03226/) (2024)
- [T*: Re-thinking Temporal Search for Long-Form Video Understanding](https://mlsanity.com/paper/2504.02259/) (2025)
- [Too Many Frames, Not All Useful: Efficient Strategies for Long-Form Video QA](https://mlsanity.com/paper/2406.09396/) (2024)
- [KeyVideoLLM: Towards Large-scale Video Keyframe Selection](https://mlsanity.com/paper/2407.03104/) (2024)
- [MDP3: A Training-free Approach for List-wise Frame Selection in Video-LLMs](https://mlsanity.com/paper/2501.02885/) (2025)
- [Fair and Diverse DPP-based Data Summarization](https://mlsanity.com/paper/1802.04023/) (2018)
- [Unleashing Hour-Scale Video Training for Long Video-Language Understanding](https://mlsanity.com/paper/2506.05332/) (2025)
- [Discovering Spatio-Temporal Rationales for Video Question Answering](https://mlsanity.com/paper/2307.12058/) (2023)

## Built on by (most cited first, 20 of 20)

- [EVA: Efficient Reinforcement Learning for End-to-End Video Agent](https://mlsanity.com/paper/2603.22918/) (2026)
- [A Benchmark for Omni-Modal Reasoning in Long Videos](https://mlsanity.com/paper/2512.16978/) (2025)
- [Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval](https://mlsanity.com/paper/2605.16481/) (2026)
- [EvoVid: Temporal-Centric Self-Evolution for Video Large Language Models](https://mlsanity.com/paper/2605.21931/) (2026)
- [OmniReasoner: Thinking with Long Audio-Video via Native Tool Use](https://mlsanity.com/paper/2607.19339/) (2026)
- [AdaptToken: Entropy-based Adaptive Token Selection for MLLM Long Video Understanding](https://mlsanity.com/paper/2603.28696/) (2026)
- [Reflect-R1: Evidence-Driven Reflection for Self-Correction in Long Video Understanding](https://mlsanity.com/paper/2606.27922/) (2026)
- [Incentivizing Vision Language Models to Search for Long Video Question Answering](https://mlsanity.com/paper/2607.02959/) (2026)
- [SEED: Targeted Data Selection by Weighted Independent Set](https://mlsanity.com/paper/2605.15691/) (2026)
- [Decomposing Queries into Tool Calls for Long-Video Keyframe Retrieval](https://mlsanity.com/paper/2605.23826/) (2026)
- [SVI-Bench: A Dynamic Microworld for Strategic Video Intelligence](https://mlsanity.com/paper/2605.31529/) (2026)
- [VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context](https://mlsanity.com/paper/2606.30288/) (2026)
- [QSVideo: Query-Conditioned Semantic Temporal Retrieval for Video Understanding](https://mlsanity.com/paper/2607.04559/) (2026)
- [Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA](https://mlsanity.com/paper/2607.16189/) (2026)
- [CADER: Confidence-Aware Dynamic Evidence Reasoning for Long-Video Understanding](https://mlsanity.com/paper/2607.24582/) (2026)
- [LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception](https://mlsanity.com/paper/2609.39938/) (2026)
- [FORTE: Adaptive Scoring and Exact Keyframe Selection for Long-Video Question Answering](https://mlsanity.com/paper/2610.00573/) (2026)
- [CASE: Cost-Aware Stopping for Efficient Long-Video Agents](https://mlsanity.com/paper/2610.05400/) (2026)
- [VAMR: Multi-Question Agentic Reasoning for Efficient Long-Form Video Understanding](https://mlsanity.com/paper/2610.11171/) (2026)
- [VideoEvoHarness: Self Evolving Agent Harness from Unlabeled Videos for Video Understanding](https://mlsanity.com/paper/2610.15389/) (2026)

## Abstract (arXiv)

Temporal search aims to identify a minimal set of relevant frames from tens of thousands based on a given query, serving as a foundation for accurate long-form video understanding. Existing works attempt to progressively narrow the search space. However, these approaches typically rely on a hand-crafted search process, lacking end-to-end optimization for learning optimal search strategies. In this paper, we propose TimeSearch-R, which reformulates temporal search as interleaved text-video thinking, seamlessly integrating searching video clips into the reasoning process through reinforcement learning (RL). However, applying RL training methods, such as Group Relative Policy Optimization (GRPO), to video reasoning can result in unsupervised intermediate search decisions. This leads to insufficient exploration of the video content and inconsistent logical reasoning. To address these issues, we introduce GRPO with Completeness Self-Verification (GRPO-CSV), which gathers searched video frames from the interleaved reasoning process and utilizes the same policy model to verify the adequacy of searched frames, thereby improving the completeness of video reasoning. Additionally, we construct datasets specifically designed for the SFT cold-start and RL training of GRPO-CSV, filtering out samples with weak temporal dependencies to enhance task difficulty and improve temporal search capabilities. Extensive experiments demonstrate that TimeSearch-R achieves significant improvements on temporal search benchmarks such as Haystack-LVBench and Haystack-Ego4D, as well as long-form video understanding benchmarks like VideoMME and MLVU. Notably, TimeSearch-R establishes a new state-of-the-art on LongVideoBench with 4.1% improvement over the base model Qwen2.5-VL and 2.0% over the advanced video reasoning model Video-R1. Our code is available at https://github.com/Time-Search/TimeSearch-R.