SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.19686 · 2026

2 RELATED WORK Reinforcement learning has emerged as a powerful paradigm for enhancing the reasoning abilities of LLMs

Qi She, Hao Zhang, Xudong Jiang, Sheng Jin, Zhongrong Zuo, Jiawei Wu, Han Qiu, Ziyue Wang

arXiv · PDF · Open in the Atlas

Code that ran

8 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 4 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2601.19686.json · badge: /api/ran/2601.19686.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
zywang0104/Video-KTR canonical (classifier label)
classifier basis: census
8 ran · 4 no recorded run
FunctionStatusWhere it lives
ceil_by_factor Ran
ran, honoured the contract we drafted
zywang0104/Video-KTR/src/qwen-vl-utils/src/qwen_vl_utils/vision_process.py
pointer only (licence: NONE) · get_code("6e45201fa27cb24a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_problem_solution Ran
ran; our contract draft was wrong, not the code
zywang0104/Video-KTR/src/r1-v/local_scripts/prepare_hf_data.py
pointer only (licence: NONE); the stored copy get_code serves, from AMAP-ML/Q-Hawkeye, is pointer only (licence: NONE) · get_code("41c548a25e90eb39")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_alternative Ran
ran; our contract draft was wrong, not the code
zywang0104/Video-KTR/src/r1-v/Evaluation/check_path_mp4.py
pointer only (licence: NONE) · get_code("267b2cec67ab7205")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

floor_by_factor Ran
ran, honoured the contract we drafted
zywang0104/Video-KTR/src/qwen-vl-utils/src/qwen_vl_utils/vision_process.py
pointer only (licence: NONE) · get_code("8155263d7ff19bb3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

format_reward Ran
ran, honoured the contract we drafted
zywang0104/Video-KTR/src/r1-v/src/open_r1/grpo.py
pointer only (licence: NONE) · get_code("26720b6f9ee7eba4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_image_data_url Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zywang0104/Video-KTR/src/r1-v/local_scripts/create_vision_cot_data.py
pointer only (licence: NONE); the stored copy get_code serves, from minglllli/cls-rl, is pointer only (licence: NONE) · get_code("3f31b48fd862935d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

prepare_dataset Ran
ran; our contract draft was wrong, not the code
zywang0104/Video-KTR/src/r1-v/src/open_r1/sft_video.py
pointer only (licence: NONE) · get_code("69e027783b4d6f9f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

round_by_factor Ran
ran, honoured the contract we drafted
zywang0104/Video-KTR/src/qwen-vl-utils/src/qwen_vl_utils/vision_process.py
pointer only (licence: NONE) · get_code("e252767324188623")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

convert_to_mp4 Not yet run
unverified (harvested, no recorded run)
zywang0104/Video-KTR/src/r1-v/Evaluation/check_path_mp4.py
pointer only (licence: NONE) · get_code("f10ba5e04cadc46e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

download_video Not yet run
unverified (harvested, no recorded run)
zywang0104/Video-KTR/src/r1-v/src/open_r1/sft_video.py
pointer only (licence: NONE) · get_code("9de1d57aec154c7d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

is_strict_mp4 Not yet run
unverified (harvested, no recorded run)
zywang0104/Video-KTR/src/r1-v/Evaluation/check_path_mp4.py
pointer only (licence: NONE) · get_code("e79f4f936da3076c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

load_image_from_path Not yet run
unverified (harvested, no recorded run)
zywang0104/Video-KTR/src/r1-v/local_scripts/prepare_hf_data.py
pointer only (licence: NONE); the stored copy get_code serves, from AMAP-ML/Q-Hawkeye, is pointer only (licence: NONE) · get_code("bdcc94560fb87f30")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zywang0104/Video-KTR when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 6 the same month, 3 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
round_by_factor 25 other papers
21 dated earlier, 1 the same month, 2 later, 1 undated
2409.12191 (2024-09)
Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
2410.11623 (2024-10)
VidEgoThink: Assessing Egocentric Video Understanding Capabilities for Embodied AI
2412.01720 (2024-12)
LamRA: Large Multimodal Model as Your Advanced Retrieval Assistant
and 22 more not named here
ceil_by_factor 23 other papers
19 dated earlier, 1 the same month, 2 later, 1 undated
2409.12191 (2024-09)
Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
2410.11623 (2024-10)
VidEgoThink: Assessing Egocentric Video Understanding Capabilities for Embodied AI
2412.01720 (2024-12)
LamRA: Large Multimodal Model as Your Advanced Retrieval Assistant
and 20 more not named here
floor_by_factor 23 other papers
19 dated earlier, 1 the same month, 2 later, 1 undated
2409.12191 (2024-09)
Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
2410.11623 (2024-10)
VidEgoThink: Assessing Egocentric Video Understanding Capabilities for Embodied AI
2412.01720 (2024-12)
LamRA: Large Multimodal Model as Your Advanced Retrieval Assistant
and 20 more not named here
get_image_data_url 8 other papers
6 dated earlier, 2 the same month
2503.16188 (2025-03)
To Think or Not To Think: A Study of Thinking in Rule-Based Visual Reinforcement Fine-Tuning
2503.21620 (2025-03)
UI-R1: Enhancing Efficient Action Prediction of GUI Agents by Reinforcement Learning
2505.14460 (2025-05)
VisualQuality-R1: Reasoning-Induced Image Quality Assessment via Reinforcement Learning to Rank
and 5 more not named here
extract_problem_solution 7 other papers
5 dated earlier, 2 the same month
2503.21620 (2025-03)
UI-R1: Enhancing Efficient Action Prediction of GUI Agents by Reinforcement Learning
2505.14460 (2025-05)
VisualQuality-R1: Reasoning-Induced Image Quality Assessment via Reinforcement Learning to Rank
2505.15810 (2025-05)
GUI-G1: Understanding R1-Zero-Like Training for Visual Grounding in GUI Agents
and 4 more not named here
load_image_from_path 7 other papers
5 dated earlier, 2 the same month
2503.21620 (2025-03)
UI-R1: Enhancing Efficient Action Prediction of GUI Agents by Reinforcement Learning
2505.14460 (2025-05)
VisualQuality-R1: Reasoning-Induced Image Quality Assessment via Reinforcement Learning to Rank
2505.15810 (2025-05)
GUI-G1: Understanding R1-Zero-Like Training for Visual Grounding in GUI Agents
and 4 more not named here
format_reward 3 other papers
3 dated earlier
2503.13377 (2025-03)
Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding
2505.12434 (2025-05)
VIDEORFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning
2510.23569 (2025-10)
EgoThinker: Unveiling Egocentric Reasoning with Spatio-Temporal CoT
download_video 1 other paper
1 dated earlier
2505.12434 (2025-05)
VIDEORFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Reinforcement learning (RL) has shown strong potential for enhancing reasoning in multimodal large language models, yet existing video reasoning methods often rely on coarse sequence-level rewards or single-factor token selection, neglecting fine-grained links among visual inputs, temporal dynamics, and linguistic outputs, limiting both accuracy and interpretability. We propose Video-KTR, a modality-aware policy shaping framework that performs selective, token-level RL by combining three attribution signals: (1) visual-aware tokens identified via counterfactual masking to reveal perceptual dependence; (2) temporal-aware tokens detected through frame shuffling to expose temporal sensitivity; and (3) high-entropy tokens signaling predictive uncertainty. By reinforcing only these key tokens, Video-KTR focuses learning on semantically informative, modality-sensitive content while filtering out low-value tokens. Across five challenging benchmarks, Video-KTR achieves state-of-the-art or highly competitive results-42.7% on Video-Holmes (surpassing GPT-4o)-with consistent gains on both reasoning and general video understanding tasks. Ablation studies verify the complementary roles of the attribution signals and the robustness of targeted token-level updates. Overall, Video-KTR improves accuracy and interpretability, offering a simple, drop-in extension to RL for complex video reasoning. Our code and models are available at https://github.com/zywang0104/Video-KTR. Reinforcement learning (RL) has emerged as a powerful paradigm for enhancing long-chain reasoning in large language models (LLMs) (OpenAI, 2024;DeepSeek-AI, 2025;Yang et al., 2025). In particular, GRPO (Shao et al., 2024) exhibits strong performance on complex reasoning tasks. Building on this success, RL has been extended to multimodal LLMs (MLLMs), achieving notable gains on visual understanding through sample diversification (

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2601.19686")
get_code_for_paper("2601.19686")
have("2601.19686")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2601.19686.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2601.19686.svg)](https://syntology.ai/paper/2601.19686)

Connect an agent — have() is free.