SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.02258 · 2026

Alignment-Aware Model Adaptation via Feedback-Guided Optimization

Leonid Sigal, Gaurav Bhatt, Jiawei Zhou, Aditya Chinchure

arXiv · PDF · Open in the Atlas

Code that ran

11 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; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 0 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/2602.02258.json · badge: /api/ran/2602.02258.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
facebookresearch/TruthRL canonical (classifier label)
classifier basis: census
11 ran · 0 no recorded run
FunctionStatusWhere it lives
em_check Ran
ran, honoured the contract we drafted
facebookresearch/TruthRL/training/verl/verl/utils/reward_score/truthrl_qa.py
pointer only (licence: NONE); the stored copy get_code serves, from TencentBAC/ReSeek, is pointer only (licence: NONE) · get_code("f8ea5de03c824510")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, facebookresearch/TruthRL 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_OOK_questions Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/data_utils/find_OOK_questions.py
pointer only (licence: NOASSERTION) · get_code("26c2b136346eb478")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/TruthRL (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.

format_reward Ran
ran, honoured the contract we drafted
facebookresearch/TruthRL/training/open-r1/src/open_r1/rewards.py
pointer only (licence: NONE) · get_code("ce353d566f3f2c9f")
How it ran

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

load_docs Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/data_utils/retrieval.py
pointer only (licence: NONE); the stored copy get_code serves, from tengxiao1/MR-Search, is pointer only (licence: NONE) · get_code("38fb2b7a0a23bbe3")
How it ran

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

load_results Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/data_utils/find_OOK_questions.py
pointer only (licence: NOASSERTION) · get_code("0a937568fc0b343b")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/TruthRL (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.

normalize_answer Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/evaluation/evaluate.py
pointer only (licence: NONE); the stored copy get_code serves, from shmsw25/GraphRetriever, is pointer only (licence: NONE) · get_code("dae7ab386661a4f4")
How it ran

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

parse_response Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/evaluation/evaluate.py
pointer only (licence: NOASSERTION) · get_code("21fa52735b227537")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/TruthRL (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.

read_jsonl Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/data_utils/retrieval.py
pointer only (licence: NONE); the stored copy get_code serves, from tengxiao1/MR-Search, is pointer only (licence: NONE) · get_code("e50ece300cf79a17")
How it ran

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

subem_check Ran
ran, honoured the contract we drafted
facebookresearch/TruthRL/training/verl/verl/utils/reward_score/truthrl_qa.py
pointer only (licence: NONE) · get_code("eb94037d0866de26")
How it ran

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

tag_count_reward Ran
ran, honoured the contract we drafted
facebookresearch/TruthRL/training/open-r1/src/open_r1/rewards.py
pointer only (licence: NONE) · get_code("0a5673b11c91e69d")
How it ran

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

truncate_prompt Ran
ran; our contract draft was wrong, not the code
facebookresearch/TruthRL/training/open-r1/src/open_r1/sft.py
pointer only (licence: NOASSERTION) · get_code("801fe22f00cc39bd")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/TruthRL (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.

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 (7 with at least one paper dated earlier, 5 the same month, 7 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
normalize_answer 133 other papers
120 dated earlier, 4 the same month, 8 later, 1 undated · 1 only as an assembly brick
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 130 more not named here
em_check 5 other papers
1 dated earlier, 2 the same month, 2 later
2510.00568 (2025-10)
ReSeek: A Self-Correcting Framework for Search Agents with Instructive Rewards
2602.02518 (2026-02)
GRAPHDANCER: Training LLMs to Explore and Reason over Graphs via Two-Stage Curriculum Post-Training
2602.03645 (2026-02)
Reinforcement Fine-Tuning for History-Aware Dense Retriever in RAG
and 2 more not named here
format_reward 5 other papers
3 dated earlier, 1 the same month, 1 later
2506.20639 (2025-06)
DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation
2507.02834 (2025-07)
ExPO: Unlocking Hard Reasoning with Self-Explanation-Guided Reinforcement Learning
2511.04902 (2025-11)
You Need Reasoning to Learn Reasoning: The Limitations of Label-Free RL in Weak Base Models
and 2 more not named here
tag_count_reward 4 other papers
2 dated earlier, 1 the same month, 1 later
2506.20639 (2025-06)
DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation
2507.02834 (2025-07)
ExPO: Unlocking Hard Reasoning with Self-Explanation-Guided Reinforcement Learning
2602.03876 (2026-02)
GOPO: Policy Optimization using Ranked Rewards
and 1 more not named here
load_docs 3 other papers
2 dated earlier, 1 later
2505.17612 (2025-05)
Distilling LLM Agent into Small Models with Retrieval and Code Tools
2601.14896 (2026-01)
Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation
2603.11327 (2026-03)
Meta-Reinforcement Learning with Self-Reflection for Agentic Search
subem_check 3 other papers
1 the same month, 2 later
2602.02518 (2026-02)
GRAPHDANCER: Training LLMs to Explore and Reason over Graphs via Two-Stage Curriculum Post-Training
2603.11327 (2026-03)
Meta-Reinforcement Learning with Self-Reflection for Agentic Search
2608.01904 (2026-08)
CoEvoKG: Co-Evolving Knowledge Graphs with Self-Evolving Search Agents
parse_response 2 other papers
2 dated earlier
2406.04744 (2024-06)
CRAG -Comprehensive RAG Benchmark
2601.11443 (2026-01)
Predict the Retrieval! Test time adaptation for Retrieval Augmented Generation
read_jsonl 2 other papers
1 dated earlier, 1 later
2601.14896 (2026-01)
Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation
2603.11327 (2026-03)
Meta-Reinforcement Learning with Self-Reflection for Agentic Search

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

Fine-tuning is the primary mechanism for adapting foundation models to downstream tasks; however, standard approaches largely optimize task objectives in isolation and do not account for secondary yet critical alignment objectives (e.g., safety and hallucination avoidance). As a result, downstream fine-tuning can degrade alignment and fail to correct pre-existing misaligned behavior. We propose an alignment-aware fine-tuning framework that integrates feedback from an external alignment signal through policygradient-based regularization. Our method introduces an adaptive gating mechanism that dynamically balances supervised and alignment-driven gradients on a per-sample basis, prioritizing uncertain or misaligned cases while allowing wellaligned examples to follow standard supervised updates. The framework further learns abstention behavior for fully misaligned inputs, incorporating conservative responses directly into the finetuned model. Experiments on general and domainspecific instruction-tuning benchmarks demonstrate consistent reductions in harmful and hallucinated outputs without sacrificing downstream task performance. Additional analyses show robustness to adversarial fine-tuning, prompt-based attacks, and unsafe initializations, establishing adaptively gated alignment optimization as an effective approach for alignment-preserving and alignment-recovering model adaptation.

For agents

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

get_harvested_code_for_paper("2602.02258")
get_code_for_paper("2602.02258")
have("2602.02258")

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

curl https://syntology.ai/api/ran/2602.02258.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.02258.svg)](https://syntology.ai/paper/2602.02258)

Connect an agent — have() is free.