SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.03608 · 2026

Exploiting Verification-Generation Gap: Test-Time Reinforcement Learning with Confidence-Conditioned Verification

Wenjie Feng, See-Kiong Ng, Jian Lou, Shunyu Wu, Dan Li, Jiahui Li, Jianfeng Shan, Wenpei Chen

arXiv · PDF

Code that ran

2 of 11 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 9 no recorded run), measured 2026-11-21; 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/2606.03608.json · badge: /api/ran/2606.03608.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
shanjf666/CoCoV canonical (classifier label)
classifier basis: census
2 ran · 9 no recorded run
FunctionStatusWhere it lives
init_mcore_model_qwen2_moe Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
shanjf666/CoCoV/verl/verl/models/mcore/model_initializer.py
pointer only (licence: NONE) · get_code("db6a3399cbd9ff7d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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.

unpad_dataproto Ran
ran; our contract draft was wrong, not the code
shanjf666/CoCoV/verl/verl/protocol.py
pointer only (licence: NONE) · get_code("25f0ea3f460f6ce1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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_generation_config Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/utils/model.py
pointer only (licence: NONE) · get_code("68aa34a57b353cae")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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_huggingface_actor_config Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/utils/model.py
pointer only (licence: NONE) · get_code("9411d1fb7e4f68e5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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_weight_loader Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/models/weight_loader_registry.py
pointer only (licence: NONE) · get_code("acf240079d434746")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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_weight_saver Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/models/weight_loader_registry.py
pointer only (licence: NONE) · get_code("d1e077d222a806ec")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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.

init_mcore_model Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/models/mcore/registry.py
pointer only (licence: NONE) · get_code("7352f43d0a5f4491")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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.

init_mcore_model_dense Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/models/mcore/model_initializer.py
pointer only (licence: NONE) · get_code("c721e4bc593beb04")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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.

init_mcore_model_llama4 Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/models/mcore/model_initializer.py
pointer only (licence: NONE) · get_code("9547c4285fb6502b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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_state_dict_to_megatron_gptmodel Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/models/mcore/loader.py
pointer only (licence: NONE) · get_code("0d364d43436cf1dd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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.

squeeze Not yet run
unverified (harvested, no recorded run)
shanjf666/CoCoV/verl/verl/utils/model.py
pointer only (licence: NONE) · get_code("3b15e2ac7497c441")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shanjf666/CoCoV 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 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 5 the same month, 4 later) and 3 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
unpad_dataproto 89 other papers
71 dated earlier, 7 the same month, 8 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.21236 (2024-10)
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models
2025.findings-emnlp.1015 (2025 (year only))
MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learning
and 86 more not named here
squeeze 82 other papers
66 dated earlier, 6 the same month, 7 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.21236 (2024-10)
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models
2025.findings-emnlp.1015 (2025 (year only))
MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learning
and 79 more not named here
get_generation_config 63 other papers
50 dated earlier, 4 the same month, 7 later, 2 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.21236 (2024-10)
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models
2502.19613 (2025-02)
Self-rewarding correction for mathematical reasoning
and 60 more not named here
get_weight_loader 58 other papers
42 dated earlier, 6 the same month, 8 later, 2 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2505.14362 (2025-05)
DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning
2505.20046 (2025-05)
REARANK: Reasoning Re-ranking Agent via Reinforcement Learning
and 55 more not named here
get_huggingface_actor_config 53 other papers
47 dated earlier, 4 the same month, 2 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.21236 (2024-10)
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models
2025.findings-emnlp.1015 (2025 (year only))
MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learning
and 50 more not named here
get_weight_saver 2 other papers
2 dated earlier
2505.20046 (2025-05)
REARANK: Reasoning Re-ranking Agent via Reinforcement Learning
2509.25787 (2025-09)
Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Ranking
init_mcore_model 1 other paper
1 dated earlier
2603.05308 (2026-03)
Med-V1: Small Language Models for Zero-shot and Scalable Biomedical Evidence Attribution
load_state_dict_to_megatron_gptmodel 1 other paper
1 dated earlier
2603.05308 (2026-03)
Med-V1: Small Language Models for Zero-shot and Scalable Biomedical Evidence Attribution

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Test-time reinforcement learning (TTRL) has emerged as a promising paradigm for enhancing the complex reasoning abilities of large language models (LLMs) in a completely label-free manner. Despite existing studies focusing on Pass@1 performance, optimizing Pass@k remains under-explored yet critical in label-free settings, which measures generation coverage for sustained exploration. Optimizing Pass@k in label-free setting is highly non-trivial, as directly applying the Pass@k advantage designs effective for RLVR yields unsatisfactory performance. Through in-depth empirical analysis, we discover the root causes hindering performance: pseudo-label estimations for low-confidence samples have a high probability of being incorrect, while candidate answers for high-confidence samples suffer from severe diversity collapse. To overcome these hurdles, we propose TTRL-CoCoV (Test-Time Reinforcement Learning with Confidence-Conditioned Verification), a novel confidence-adaptive framework that expands Pass@k coverage and improves Pass@1 performance. Based on our key insight that verification capability generally leads generation capability, TTRL-CoCoV employs a confidence-conditioned mechanism: for high-confidence samples, it bootstraps verifier and applies an exploration-enhancing reward to prevent diversity collapse; for low-confidence samples, it delegates pseudo-label selection to the verifier to filter incorrect pseudolabels; and for medium-confidence samples, it bypasses verification entirely. Extensive experiments demonstrate that TTRL-CoCoV outperforms the best competing methods across 6 widely-recognized benchmarks, achieves average absolute gains of +9.8% in Pass@1 and +18.7% in Pass@16 over TTRL, and even achieves absolute Pass@1 improvements of up to +5.0% across multiple reasoning benchmarks when compared against fully supervised RL methods. Our code repository: https://github.com/shanjf666/CoCoV.

For agents

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

get_harvested_code_for_paper("2606.03608")
get_code_for_paper("2606.03608")
have("2606.03608")

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

curl https://syntology.ai/api/ran/2606.03608.json

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

[![Syntology run record](https://syntology.ai/api/ran/2606.03608.svg)](https://syntology.ai/paper/2606.03608)

Connect an agent — have() is free.