Supervised Fine-Tuning

Follow this topic Ask about this topic

Recent momentum

+30%

13 papers in the last 28 days · 0.2% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

Weekly history

Recent digests

What was published in this topic, kept on the site without email delivery.

Period ending 2026-09-21

4 new papers

A weekly snapshot of new work published in Supervised Fine-Tuning.

Period ending 2026-09-14

1 new paper

A weekly snapshot of new work published in Supervised Fine-Tuning.

Period ending 2026-09-07

7 new papers

A weekly snapshot of new work published in Supervised Fine-Tuning.

109 papers

Latest in Supervised Fine-Tuning

Browse all topics →

Post-Training is About States, Not Tokens: A State Distribution View of SFT, RL, and On-Policy Distillation
May 21, 2026Dong NieLLM Post-Training Methods Supervised Fine-Tuning
From Patches to Trajectories: Privileged Process Supervision for Software-Engineering Agents
May 21, 2026Murong Ma, Tianyu Chen, Yun Lin +7Swe-Bench Verified Supervised Fine-Tuning
Models Can Model, But Can't Bind: Structured Grounding in Text-to-Optimization
May 20, 2026Zhiqi Gao, Albert Ge, Alexander Berenbeim +2Supervised Fine-Tuning Longbench
Thinking-while-speaking: A Controlled, Interleaved Reasoning Method for Real-Time Speech Generation
May 20, 2026Xuan Du, Qiangyu Yan, Wenshuo Li +4Speech-Llms Supervised Fine-Tuning
Pareto-Enhanced Portrait Generation: Vision-Aligned Text Supervision for Alignment, Realism, and Aesthetics
May 20, 2026Yunlong Wang, Jinjin Shi, Wenbin Gao +3Text-To-Image Diffusion Models Video-Text Alignment
Reconciling Contradictory Views on the Effectiveness of SFT in LLMs: An Interaction Perspective
May 18, 2026Junpeng Zhang, Lei Cheng, Guoxi Zhang +3Supervised Fine-Tuning Overfitting
SNLP: Layer-Parallel Inference via Structured Newton Corrections
May 18, 2026Ligong Han, Kai Xu, Hao Wang +1Transformer Layers Transformer Residual Streams
ClaHF: A Human Feedback-inspired Reinforcement Learning Framework for Improving Classification Tasks
May 17, 2026Tianxiang Xu, Xiaoyan Zhu, Xin Lai +1Reinforcement Learning From Human Feedback Supervised Fine-Tuning
MixSD: Mixed Contextual Self-Distillation for Knowledge Injection
May 16, 2026Jiarui Liu, Lechen Zhang, Yongjin Yang +5Knowledge Distillation Supervised Fine-Tuning
Sparse Autoencoders enable Robust and Interpretable Fine-tuning of CLIP models
May 15, 2026Fabian Morelli, Arnas Uselis, Ankit Sonthalia +1Supervised Fine-Tuning Large Vision Language Models
Boosting Reinforcement Learning with Verifiable Rewards via Randomly Selected Few-Shot Guidance
May 14, 2026Kai Yan, Alexander G. Schwing, Yu-Xiong WangReinforcement Learning With Verifiable Reward Verifiable Rewards
InfoSFT: Learn More and Forget Less with Information-Aware Token Weighting
May 14, 2026Mahdi Sabbaghi, George Pappas, Adel Javanmard +1Supervised Fine-Tuning
Reducing the Safety Tax in LLM Safety Alignment with On-Policy Self-Distillation
May 14, 2026Yu Fu, Longxuan Yu, Haz Sameen Shahgir +4Unsupervised On-Policy Self-Distillation Safety Alignment
Children's English Reading Story Generation via Supervised Fine-Tuning of Compact LLMs with Controllable Difficulty and Safety
May 13, 2026Qian Shen, Fanghua Cao, Min Yao +3Llm-Generated Stories Large Language
From Instance Selection to Fixed-Pool Data Recipe Search for Supervised Fine-Tuning
May 13, 2026Haodong Wu, Jiahao Zhang, Lijie Hu +1Supervised Fine-Tuning Table Reasoning Datasets
Data Difficulty and the Generalization--Extrapolation Tradeoff in LLM Fine-Tuning
May 13, 2026Siyuan Liu, Tinghong Chen, Xinghan Li +2Large Language Model Fine-Tuning Supervised Fine-Tuning
Reconstruction of Personally Identifiable Information from Supervised Finetuned Models
May 12, 2026Sae Furukawa, Alina OpreaSupervised Fine-Tuning Language Model Evasion Attacks
A Mechanistic Investigation of Supervised Fine Tuning
May 12, 2026Ruhaan ChopraSupervised Fine-Tuning Model Activations
CapVector: Learning Transferable Capability Vectors in Parametric Space for Vision-Language-Action Models
May 11, 2026Wenxuan Song, Han Zhao, Fuhao Li +7Vision-Language Model Adaptation Vision-Language-Action Models
Memorize Theorems, Not Instances: Probing SFT Generalization through Mathematical Reasoning
May 10, 2026Ruiying Peng, Mengyu Yang, Jing Lei +3Supervised Fine-Tuning Mathematical Reasoning
SearchSkill: Teaching LLMs to Use Search Tools with Evolving Skill Banks
May 9, 2026Jinchao Hu, Meizhi Zhong, Kehai Chen +1Agentic Search Multi-Hop Question Answering Benchmarks
Large Language Models for Sequential Decision-Making: Improving In-Context Learning via Supervised Fine-Tuning
May 9, 2026Minmin Zhang, Sina Aghaei, Soroush SaghafianIn-Context Learning Large Language Model Fine-Tuning
Preserving Foundational Capabilities in Flow-Matching VLAs through Conservative SFT
May 9, 2026Tianyi Zhang, Shaopeng Zhai, Haoran Zhang +2Flow-Matching Vision-Language-Action Supervised Fine-Tuning
Rotation-Preserving Supervised Fine-Tuning
May 8, 2026Hangzhan Jin, Tianwei Ni, Lu Li +3Supervised Fine-Tuning Fisher Information
Optimizer-Model Consistency: Full Finetuning with the Same Optimizer as Pretraining Forgets Less
May 7, 2026Yuxing Liu, Jianyu Wang, Tong ZhangSupervised Fine-Tuning Model Pretraining
Crafting Reversible SFT Behaviors in Large Language Models
May 7, 2026Yuping Lin, Pengfei He, Yue Xing +5Supervised Fine-Tuning Model Activations
Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
May 7, 2026Oğuzhan Fatih Kar, Roman Bachmann, Yuanzheng Gong +2Web Agents Web
D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models
May 6, 2026Dengyang Jiang, Xin Jin, Dongyang Liu +9On-Policy Distillation Rp-Opsd
Stabilizing LLM Supervised Fine-Tuning via Explicit Distributional Control
May 6, 2026Xinyu Wang, Changzhi Sun, Yuanbin Wu +1Supervised Fine-Tuning Finetuning
Reference-Sampled Boltzmann Projection for KL-Regularized RLVR: Target-Matched Weighted SFT, Finite One-Shot Gaps, and Policy Mirror Descent
May 4, 2026Yao Shu, Chenxing Wei, Hongbin Lin +2Reinforcement Learning With Verifiable Reward Verifiable Rewards
Decouple before Integration: Test-time Synthesis of SFT and RLVR Task Vectors
May 1, 2026Chaohao Yuan, Chenghao Xiao, Yu Rong +2Supervised Fine-Tuning Reinforcement Learning With Verifiable Reward
Diversity in Large Language Models under Supervised Fine-Tuning
Apr 30, 2026Roman Klypa, Oleksandr CherednichenkoSupervised Fine-Tuning Diversity
Beyond SFT-to-RL: Pre-alignment via Black-Box On-Policy Distillation for Multimodal RL
Apr 30, 2026Sudong Wang, Weiquan Huang, Xiaomin Yu +9Multimodal Reasoning Supervised Fine-Tuning
How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum
Apr 28, 2026Chu-Cheng Lin, Eugene IeSupervised Fine-Tuning Large Reasoning Models
Compute Aligned Training: Optimizing for Test Time Inference
Apr 27, 2026Adam Ousherovitch, Ambuj TewariTest-Time Scaling Large Models
Reinforcement Learning Improves LLM Accuracy and Reasoning in Disease Classification from Radiology Reports
Apr 21, 2026Yishu Wei, Yi Lin, Adam Flanders +2Radiology Report Generation Disease Detection
Generative Texture Filtering
Apr 21, 2026Rongjia Zheng, Shangwei Huang, Lei Zhu +2Model Pretraining Supervised Fine-Tuning
Aligning Language Models for Lyric-to-Melody Generation with Rule-Based Musical Constraints
Apr 20, 2026Hao Meng, Siyuan Zheng, Shuran Zhou +2Music Generation Large Language Model Alignment
REALM: Reliable Expertise-Aware Language Model Fine-Tuning from Noisy Annotations
Apr 19, 2026Sajjad Ghiasvand, Mark Beliaev, Mahnoosh Alizadeh +1Large Language Model Fine-Tuning Supervised Fine-Tuning
D-QRELO: Training- and Data-Free Delta Compression for Large Language Models via Quantization and Residual Low-Rank Approximation
Apr 18, 2026Junlin Li, Shuangyong Song, Guodong Du +6Large Language Model Compression Quantization
Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Mar 3, 2026Haoran Lu, Shang Wu, Songling Liu +94D Generation Video World Models
FAIL: Flow Matching Adversarial Imitation Learning for Image Generation
Feb 12, 2026Yeyao Ma, Chen Li, Xiaosong Zhang +2Imitation Learning Image Generation
Patch the Distribution Mismatch: RL Rewriting Agent for Stable Off-Policy SFT
Feb 11, 2026Jiacheng Wang, Zhijie Liu, Ping Jian +4Supervised Fine-Tuning Rewrite
Entropy-Preserving Supervised Fine-Tuning via Adaptive Self-Distillation for Large Reasoning Models
Feb 2, 2026Hao Wang, Hao Gu, Hongming Piao +6Supervised Fine-Tuning Large Reasoning Models
Self-Improvement as Coherence Optimization: A Theoretical Account
Jan 20, 2026Tianyi Qiu, Ahmed Hani Ismail, Zhonghao He +1Supervised Fine-Tuning Self-Training
Solving Semi-Supervised Few-Shot Learning from an Auto-Annotation Perspective
Dec 11, 2025Tian Liu, Anwesha Basu, James Caverlee +1Few-Shot Learning Semi-Supervised Learning
RAFP: Identifying LLM Lineages via Rare-Region Fingerprints
May 19, 2025Yun-Yun Tsai, Jia Hao Liang, Chuan Guo +2Fingerprint Large Language Model Fine-Tuning
Palmyra x6 Technical Report: An Agentic, Tool-Use Model Post-Trained via Anchored Supervised Fine-Tuning
Date pendingPeng Du, Kiran Kamble, Rakshith Vasudev +5Supervised Fine-Tuning Agentic Tool Use
Complementing reinforcement learning with SFT through logit averaging in the post training of LLMs
Date pendingXingwei Gan, Ying ZhuGroup Relative Policy Optimization Reinforcement Learning Post-Training

May 21, 2026 · cs.LGJ/K move · Enter open · S save

Post-Training is About States, Not Tokens: A State Distribution View of SFT, RL, and On-Policy Distillation

Dong Nie

Independent Researcher

Large language model post-training methods such as supervised fine-tuning (SFT), reinforcement learning (RL), and distillation are often analyzed through their loss functions: maximum likelihood, policy gradients, forward KL, reverse KL, or related objective-level variants. We study a complementary factor: the state distribution on which supervision is applied. For an autoregressive policy, a state is a prompt plus generated prefix. SFT trains on fixed dataset states, while RL and on-policy distillation (OPD) train on states induced by the current learner. We formalize post-training as state-distribution shaping and run a controlled smallscale study using Qwen3-0.6B-Base on GSM8K, with TruthfulQA and MMLU as retention evaluations. Our results show three phenomena. First, a mild SFT run improves GSM8K with little forgetting, while a stress SFT run causes substantial retention loss. Second, OPD from a degraded SFT teacher surpasses that teacher on GSM8K, TruthfulQA, and MMLU, despite using the teacher as its only supervision source. Third, a lightweight on-policy RL run improves GSM8K while preserving retention. These results support a state-centric view of post-training: the source and locality of training states can be as important as the form of the supervision signal.

LLM Post-Training Methods Supervised Fine-Tuning On-Policy Distillation Loss Function

Open paper Ask about it PDF

Recent momentum

Neighbouring directions

Recent digests

4 new papers

1 new paper

7 new papers

Latest in Supervised Fine-Tuning