SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.01649 · 2026

Contribution-aware Token Compression for Efficient Video Understanding via Reinforcement Learning

Bo Zheng, Qiang Zhou, Zhibin Wang, Jun Song, Yinchao Ma, Xianing Chen, Hanqing Yang

arXiv · PDF · Open in the Atlas

Code that ran

13 of 16 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; 1 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 10 failures of our instrument; 3 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/2602.01649.json · badge: /api/ran/2602.01649.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
EvolvingLMMs-Lab/lmms-eval canonical (classifier label)
classifier basis: census
8 ran · 2 no recorded run
LivingFutureLab/CaCoVID canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
escaped_split Ran
ran; our contract draft was wrong, not the code
LivingFutureLab/CaCoVID/lmms_eval/lmms_eval/utils.py
pointer only (licence: NONE) · get_code("57e732507bad40db")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LivingFutureLab/CaCoVID 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.

escaped_split Ran
ran; our contract draft was wrong, not the code
EvolvingLMMs-Lab/lmms-eval/lmms_eval/utils.py
pointer only (licence: NOASSERTION) · get_code("29065ecbfb95eb0b")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, EvolvingLMMs-Lab/lmms-eval (NOASSERTION). 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_sample_size Ran
ran, honoured the contract we drafted
LivingFutureLab/CaCoVID/lmms_eval/lmms_eval/evaluator_utils.py
pointer only (licence: NONE) · get_code("00f5e7d9ee6636ba")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LivingFutureLab/CaCoVID 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.

hash_args Ran
ran; our contract draft was wrong, not the code
LivingFutureLab/CaCoVID/lmms_eval/lmms_eval/api/model.py
pointer only (licence: NONE) · get_code("20a7cc804eb22661")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LivingFutureLab/CaCoVID 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.

hash_string Ran
ran; our contract draft was wrong, not the code
EvolvingLMMs-Lab/lmms-eval/lmms_eval/utils.py
pointer only (licence: NONE) · get_code("1fddd33683ef553c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EvolvingLMMs-Lab/lmms-eval 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.

is_json Ran
ran, violated the contract we drafted
EvolvingLMMs-Lab/lmms-eval/lmms_eval/utils.py
pointer only (licence: NONE) · get_code("ae00474f91be5441")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EvolvingLMMs-Lab/lmms-eval 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.

is_package_available Ran
ran, violated the contract we drafted
EvolvingLMMs-Lab/lmms-eval/lmms_eval/imports.py
pointer only (licence: NOASSERTION) · get_code("4baf9c8d86f47526")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, EvolvingLMMs-Lab/lmms-eval (NOASSERTION). 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.

make_lazy_getattr Ran
ran; our contract draft was wrong, not the code
EvolvingLMMs-Lab/lmms-eval/lmms_eval/imports.py
pointer only (licence: NOASSERTION) · get_code("7e8f1163e46a3004")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, EvolvingLMMs-Lab/lmms-eval (NOASSERTION). 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.

optional_import Ran
ran; our contract draft was wrong, not the code
EvolvingLMMs-Lab/lmms-eval/lmms_eval/imports.py
pointer only (licence: NOASSERTION) · get_code("c845139271f3b6e1")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, EvolvingLMMs-Lab/lmms-eval (NOASSERTION). 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.

register_group Ran
ran; our contract draft was wrong, not the code
LivingFutureLab/CaCoVID/lmms_eval/lmms_eval/api/registry.py
pointer only (licence: NONE) · get_code("1bbf582194532683")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LivingFutureLab/CaCoVID 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.

register_task Ran
ran; our contract draft was wrong, not the code
LivingFutureLab/CaCoVID/lmms_eval/lmms_eval/api/registry.py
pointer only (licence: NONE) · get_code("6a4a70c8a645fde8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LivingFutureLab/CaCoVID 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.

remove_none_pattern Ran
ran; our contract draft was wrong, not the code
EvolvingLMMs-Lab/lmms-eval/lmms_eval/logging_utils.py
pointer only (licence: NONE) · get_code("74386ef186ff5c64")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EvolvingLMMs-Lab/lmms-eval 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.

request_caching_arg_to_dict Ran
ran; our contract draft was wrong, not the code
EvolvingLMMs-Lab/lmms-eval/lmms_eval/evaluator.py
pointer only (licence: NONE) · get_code("c44830a7722ff5f7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EvolvingLMMs-Lab/lmms-eval 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_model Not yet run
unverified (harvested, no recorded run)
LivingFutureLab/CaCoVID/lmms_eval/lmms_eval/api/registry.py
pointer only (licence: NONE) · get_code("894f96b26abd7ab3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LivingFutureLab/CaCoVID 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.

get_sample_size Not yet run
unverified (harvested, no recorded run)
EvolvingLMMs-Lab/lmms-eval/lmms_eval/evaluator_utils.py
pointer only (licence: NOASSERTION) · get_code("f827e5b576859cb9")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, EvolvingLMMs-Lab/lmms-eval (NOASSERTION). 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.

run_models Not yet run
unverified (harvested, no recorded run)
EvolvingLMMs-Lab/lmms-eval/lmms_eval/cli/models_cmd.py
pointer only (licence: NOASSERTION) · get_code("214f63d400f75b96")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, EvolvingLMMs-Lab/lmms-eval (NOASSERTION). 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 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 0 the same month, 7 later) and 6 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
hash_args 44 other papers
40 dated earlier, 3 later, 1 undated
2005.14165 (2020-05)
Language Models are Few-Shot Learners
2110.04374 (2021-10)
A Few More Examples May Be Worth Billions of Parameters
2303.17003 (2023-03)
Evaluating GPT-3.5 and GPT-4 Models on Brazilian University Admission Exams
and 41 more not named here
request_caching_arg_to_dict 17 other papers
14 dated earlier, 3 later
2005.14165 (2020-05)
Language Models are Few-Shot Learners
2405.07938 (2024-05)
EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning
2411.07176 (2024-11)
More Expressive Attention with Negative Weights
and 14 more not named here
get_sample_size 12 other papers
9 dated earlier, 3 later
2005.14165 (2020-05)
Language Models are Few-Shot Learners
2405.07938 (2024-05)
EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning
2412.14171 (2024-12)
Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces
and 9 more not named here
escaped_split 9 other papers
8 dated earlier, 1 later
2005.14165 (2020-05)
Language Models are Few-Shot Learners
2305.14493 (2023-05)
Do prompt positions really matter?
2307.16039 (2023-07)
Okapi: Instruction-tuned Large Language Models in Multiple Languages with Reinforcement Learning from Human Feedback
and 6 more not named here
remove_none_pattern 7 other papers
5 dated earlier, 2 later
2408.17267 (2024-08)
UrBench: A Comprehensive Benchmark for Evaluating Large Multimodal Models in Multi-View Urban Scenarios
2409.06851 (2024-09)
LIME: Less Is More for MLLM Evaluation
2412.14171 (2024-12)
Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces
and 4 more not named here
hash_string 2 other papers
1 dated earlier, 1 later
2512.06866 (2025-12)
Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior
2607.21595 (2026-07)
3D-Aware VLMs with Implicit and Explicit Geometries
is_json 2 other papers
1 dated earlier, 1 later
2512.06866 (2025-12)
Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior
2607.21595 (2026-07)
3D-Aware VLMs with Implicit and Explicit Geometries
get_model 1 other paper
1 dated earlier
2504.04715 (2025-04)
Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs
register_group 1 other paper
1 dated earlier
2504.04715 (2025-04)
Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs
register_task 1 other paper
1 dated earlier
2504.04715 (2025-04)
Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs

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

Video large language models have demonstrated remarkable capabilities in video understanding tasks. However, the redundancy of video tokens introduces significant computational overhead during inference, limiting their practical deployment. Many compression algorithms are proposed to prioritize retaining features with the highest attention scores to minimize perturbations in attention computations. However, the correlation between attention scores and their actual contribution to correct answers remains ambiguous. To address the above limitation, we propose a novel Contributionaware token Compression algorithm for VIDeo understanding (CaCoVID) that explicitly optimizes the token selection policy based on the contribution of tokens to correct predictions. First, we introduce a reinforcement learning-based framework that optimizes a policy network to select video token combinations with the greatest contribution to correct predictions. This paradigm shifts the focus from passive token preservation to active discovery of optimal compressed token combinations. Secondly, we propose a combinatorial policy optimization algorithm with online combination space sampling, which dramatically reduces the exploration space for video token combinations and accelerates the convergence speed of policy optimization. Extensive experiments on diverse video understanding benchmarks demonstrate the effectiveness of CaCoVID. Codes are available at https://github.com/LivingFutureLab/CaCoVID.

For agents

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

get_harvested_code_for_paper("2602.01649")
get_code_for_paper("2602.01649")
have("2602.01649")

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

curl https://syntology.ai/api/ran/2602.01649.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.01649.svg)](https://syntology.ai/paper/2602.01649)

Connect an agent — have() is free.