Post-Training

Follow this topic Ask about this topic

Momentum

73 papers in the last four weeks, up 306% on the four weeks before. 0.7% of all new papers.

Jul 13Week of Sep 28

Latest papers 297

All topics

Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression
May 9, 2026Hengyi Zhu, Zhendong Mi, Grace Li Zhang +1Large Language Model Compression Data Compression Methods
1Stevens Institute of Technology, Hoboken, USA · 2Technical University of Darmstadt, Darmstadt, Germany
On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective
May 8, 2026Yuhao Li, Shengchao LiuPost-Training Supervised Finetuning
Department of Computer Science and Engineering The Chinese University of Hong Kong
Flow-OPD: On-Policy Distillation for Flow Matching Models
May 8, 2026Zhen Fang, Wenxuan Huang, Yu Zeng +8Text-To-Image Diffusion Models Image-Text Alignment
University of Science and Technology of China · University of California, Los Angeles · 3The Chinese University of Hong Kong +1
KL for a KL: On-Policy Distillation with Control Variate Baseline
May 8, 2026Minjae Oh, Sangjun Song, Gyubin Choi +2Kullback-Leibler Divergence Post-Training
Graduate School of Data Science Seoul National University
Not All Tokens Learn Alike: Attention Entropy Reveals Heterogeneous Signals in RL Reasoning
May 8, 2026Gengyang Li, Zheng-Fan Wu, Siqi Bao +1Token-Level Entropy Post-Training
1Baidu · School of Software and Microelectronics, Peking University · School of Computer Science, Peking University
Post-training makes large language models less human-like
May 8, 2026Marcel Binz, Elif Akata, Abdullah Almaatouq +76Artificial Intelligence Personas Large Language Model Alignment
Emergent Symbolic Structure in Health Foundation Models: Extraction, Alignment, and Cross-Modal Transfer
May 8, 2026Gajendra Katuwal, Advait Koparkar, Salar Abbaspourazad +2Medical World Model Multimodal Foundation Model
Apple
Implicit Compression Regularization: Concise Reasoning via Internal Shorter Distributions in RL Post-Training
May 8, 2026Chen Wang, Hexuan Deng, Yining Zhang +5Short Reasoning Traces
College of Software, Nankai University · 2Zhongguancun Academy · 3Harbin Institute of Technology +4
Steering Visual Generation in Unified Multimodal Models with Understanding Supervision
May 7, 2026Zeyu Liu, Zanlin Ni, Yang Yue +5Multimodal Generation Multimodal Model
1Tsinghua University · 2Kolors Team, Kuaishou Technology
Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training
May 6, 2026Hengyu Shi, Tianyang Han, Peizhe Wang +3Post-Training Layer-Wise
1Independent Researcher · 2D4 Lab · 3Southeast University
RoboAlign-R1: Distilled Multimodal Reward Alignment for Robot Video World Models
May 5, 2026Hao Wu, Yuqi Li, Yuan Gao +10Video World Models Long-Horizon Inference
Tencent · Tsinghua University · Department of Computer Science, City University of New York +3
Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe
May 5, 2026Wenjin Hou, Shangpin Peng, Weinong Wang +13Uni-Opd Post-Training
1Zhejiang University · 2Shenzhen Loop Area Institute · 3LLM Department, Tencent
Sharpness-Aware Pretraining Mitigates Catastrophic Forgetting
May 4, 2026Ishaan Watts, Catherine Li, Sachin Goyal +2Catastrophic Forgetting Post-Training
Carnegie Mellon University
Technical Report: Activation Residual Hessian Quantization (ARHQ) for Low-Bit LLM Quantization
Apr 30, 2026YiFeng Wang, Zhun Sun, Keisuke SakaguchiLarge Language Model Quantization Post-Training
Graduate School of Information Sciences Tohoku University
Helpfulness Hurts: Domain-Dependent Degradation of Mid-Trained Compassion Values Under Post-Training
Apr 30, 2026Jasmine Brazilek, Juliana SeawellPost-Training Pretraining
†Compassion Aligned Machine Learning (CaML)
Backtranslation Augmented Direct Preference Optimization for Neural Machine Translation
Apr 28, 2026Mehrdad Ghassabi, Spehr Rajabi, Hamidreza Baradaran Kashani +3Neural Machine Translation Backtranslation Augmented Direct Preference Optimization
Faculty of Computer Engineering University of Isfahan Isfahan, Iran · School of Computer Science University of Windsor Windsor, Canada
A Systematic Post-Train Framework for Video Generation
Apr 28, 2026Zeyue Xue, Siming Fu, Jie Huang +9Video Generation Post-Training
The University of Hong Kong · JD Explore Academy · Tsinghua University +2
Why Does Reinforcement Learning Generalize? A Feature-Level Mechanistic Study of Post-Training in Large Language Models
Apr 27, 2026Dan Shi, Zhuowen Han, Simon Ostermann +3Post-Training Model Fine-Tuning
TJUNLP Lab, School of Computer Science and Technology, Tianjin University, China · German Research Center for Artificial Intelligence (DFKI), Saarbrücken, Germany · Saarland University, Saarbrücken, Germany
Latent Agents: A Post-Training Procedure for Internalized Multi-Agent Debate
Apr 27, 2026John Seon Keun Yi, Aaron Mueller, Dokyun LeeMulti-Agent Debate Debate
Boston University
Coverage-Based Calibration for Post-Training Quantization via Weighted Set Cover over Outlier Channels
Apr 27, 2026Ibne Farabi Shihab, Sanjeda Akter, Anuj SharmaPost-Training Quantization Post-Training
Department of Computer Science, Iowa State University · Department of Civil, Construction & Environmental Engineering, Iowa State University
Stabilizing Efficient Reasoning with Step-Level Advantage Selection
Apr 27, 2026Han Wang, Xiaodong Yu, Jialian Wu +4LLM Reasoning Strategies Step-Level Advantage Estimation
UNC Chapel Hill · Advanced Micro Devices, Inc.
Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Training
Apr 25, 2026Suning Huang, Jiaqi Shao, Ke Wang +5Generalist Vision--Language--Action Frozen Vision-Language Models
Stanford University
When Policies Cannot Be Retrained: A Unified Closed-Form View of Post-Training Steering in Offline Reinforcement Learning
Apr 23, 2026Elias Hossain, Mohammad Jahid Ibna Basher, Ivan Garibay +2Off-Policy Learning Offline Reinforcement Learning
College of Engineering and Computer Science University of Central Florida, Orlando, FL, USA
Hi-WM: Human-in-the-World-Model for Scalable Robot Post-Training
Apr 23, 2026Yaxuan Li, Zhongyi Zhou, Yefei Chen +5Efficient World-Action Model Scalable Robot Learning
Current Robotics · Tsinghua University · Peking University +1
ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control
Apr 22, 2026Shelly Golan, Michael Finkelson, Ariel Bereslavsky +2Multi-Objective Reinforcement Learning Diffusion Models
Tel Aviv University · Lightricks · Adobe Research
Bootstrapping Post-training Signals for Open-ended Tasks via Rubric-based Self-play on Pre-training Text
Apr 21, 2026Chengyu Huang, Sheng-Yen Chou, Zhengxin Zhang +1Self-Play Post-Training
Department of Computer Science, Cornell University
HP-Edit: A Human-Preference Post-Training Framework for Image Editing
Apr 21, 2026Fan Li, Chonghuinan Wang, Lina Lei +9Diffusion-Based Image Editing Image Editing
Huawei Noah’s Ark Lab · Nankai University · Harbin Institute of Technology
Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Training
Apr 21, 2026Yunbo Long, Tejumade Afonja, Guangya Hao +2Post-Training Next-Token Distribution
Department of Engineering, University of Cambridge · CISPA Helmholtz Center for Information Security, Saarbrücken, Germany · The Alan Turing Institute, London
Reasoning Structure Matters for Safety Alignment of Reasoning Models
Apr 21, 2026Yeonjun In, Wonjoong Kim, Sangwu Park +1Safety Alignment Large Reasoning Models
KAIST
Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts
Apr 20, 2026Jacob Morrison, Sanjay Adhikesaven, Akshita Bhagia +3Mixture-Of-Experts Post-Training
University of Washington · Allen Institute for AI · University of California, Berkeley
Post-training is (Massive) Supervised Learning
Apr 20, 2026Michael Hassid, Yossi Adi, Roy SchwartzPost-Training Pretraining
1FAIR, Meta AI · 2The Hebrew University of Jerusalem
Characterizing Model-Native Skills
Apr 19, 2026Feiyang Kang, Mahavir Dabas, Myeongseob Ko +1Skills Model Activations
Virginia Tech Blacksburg, VA 24060, USA
Step-GRPO: Internalizing Dynamic Early Exit for Efficient Reasoning
Apr 18, 2026Benteng Chen, Weida Wang, Shufei Zhang +2Large Reasoning Models Post-Training
East China Normal University · Shanghai AI Laboratory · Fudan University +1
AutoOR: Scalably Post-training LLMs to Autoformalize Operations Research Problems
Apr 18, 2026Sumeet Ramesh Motwani, Chuan Du, Aleksander Petrov +4Operations Research Auto Research
1X · University of Oxford
Where does output diversity collapse in post-training?
Apr 17, 2026Constantinos Karouzos, Xingwei Tan, Nikolaos AletrasDiversity Collapse Diversity
School of Computer Science University of Sheffield, UK
S-GRPO: Unified Post-Training for Large Vision-Language Models
Apr 17, 2026Yuming Yan, Kai Tang, Sihong Chen +4Vision-Language Model Adaptation Post-Training
Tencent Shenzhen, Guangdong, China
Discovering Novel LLM Experts via Task-Capability Coevolution
Apr 16, 2026Andrew Dai, Boris Meinardus, Ciaran Regan +2Co-Evolution Post-Training
Sakana AI
English is Not All You Need: Systematically Exploring the Role of Multilinguality in LLM Post-Training
Apr 14, 2026Mehak Dhaliwal, Shashwat Chaurasia, Yao Qin +2Multilingual Language Models Cross-Lingual Transfer
UC Santa Barbara · Amazon
PISCES: Annotation-free Text-to-Video Post-Training via Optimal Transport-Aligned Rewards
Feb 2, 2026Minh-Quan Le, Gaurav Mittal, Cheng Zhao +3Text-To-Video Generation Model Post-Training
1Microsoft · 2Stony Brook University · †Currently at Dolby Labs. 1Microsoft 2Stony Brook University
Mechanism Shift During Post-training from Autoregressive to Masked Diffusion Language Models
Jan 21, 2026Injin Kong, Hyoungjoon Lee, Yohan JoMasked Diffusion Language Models Diffusion Language Models
Graduate School of Data Science, Seoul National University · Department of Biosystems & Biomaterials Science and Engineering, Seoul National University
Value Drifts: Tracing Value Alignment During LLM Post-Training
Oct 30, 2025Mehar Bhatia, Shravan Nayak, Gaurav Kamath +4Large Language Model Alignment Preference Datasets
Mila - Quebec AI Institute · McGill University · Université de Montréal +4
A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning
Oct 21, 2025Mengqi Li, Lei Zhao, Anthony Man-Cho So +2LLM Reasoning Strategies Mathematical Reasoning Benchmarks
The Chinese University of Hong Kong, Shenzhen · Shanghai Jiao Tong University · The Chinese University of Hong Kong
Navigating the Alignment-Calibration Trade-off: A Pareto-Superior Frontier via Model Merging
Oct 20, 2025Tiancheng Hu, Benjamin Minixhofer, Nigel CollierValue Alignment Post-Training
University of Cambridge
SRUM: Fine-Grained Self-Rewarding for Unified Multimodal Models
Oct 14, 2025Weiyang Jin, Yuwei Niu, Jiaqi Liao +4Multimodal Generation Post-Training
HKU MMLab · The University of Hong Kong · Shenzhen Loop Area Institute +2
Representation-Based Exploration for Language Models: From Test-Time to Post-Training
Oct 13, 2025Jens Tuyls, Dylan J. Foster, Akshay Krishnamurthy +1Post-Training Exploration
Princeton University · Microsoft Research NYC
BaNEL: Exploration Posteriors for Generative Modeling Using Only Negative Rewards
Oct 10, 2025Sangyun Lee, Brandon Amos, Giulia FantiGenerative Models Progress Reward Modeling
Carnegie Mellon University
What Is The Political Content in LLMs' Pre- and Post-Training Data?
Sep 26, 2025Tanise Ceron, Dmitry Nikolaev, Dominik Stammbach +1Large Language Model Bias Political Discourse
Bocconi University, Italy · University of Manchester, UK · Princeton University, USA
Reconstruction Alignment Improves Unified Multimodal Models
Sep 8, 2025Ji Xie, Trevor Darrell, Luke Zettlemoyer +1Multimodal Model Post-Training
1UC Berkeley · University of Washington · 3Duke University
Post-training for Efficient Communication via Convention Formation
Aug 8, 2025Yilun Hua, Evan Wang, Yoav ArtziPost-Training Multi-Turn Interactions
Department of Computer Science and Cornell Tech Cornell University
Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training
Jul 7, 2025Song Lai, Haohan Zhao, Rong Feng +9Post-Training Reinforcement Fine-Tuning
Centre for Artificial Intelligence and Robotics, Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences, Hong Kong, China · City University of Hong Kong, Hong Kong, China · School of Artificial Intelligence, Jilin University, Changchun, China +3
Reinforcement Learning from Human Feedback
Apr 16, 2025Nathan LambertReinforcement Learning From Human Feedback Post-Training
EFC++: Elastic Feature Consolidation with Prototype Re-balancing for Cold Start Exemplar-free Incremental Learning
Mar 13, 2025Simone Magistri, Tomaso Trinci, Albin Soutif-Cormerais +2Class-Incremental Learning Learnable Prototypes
Media Integration and Communication Center (MICC), University of Florence, Italy · Global Optimization Laboratory, University of Florence, Italy · LAMP Team, Computer Vision Center, Barcelona, Spain +1
Unified Text-Image Generation with Weakness-Targeted Post-Training
Date pendingJiahui Chen, Philippe Hansen-Estruch, Xiaochuang Han +7Multimodal Generation Text-To-Image
FAIR at Meta · University of Texas, Austin · Work done at Meta +2
Benford's Law as a Distributional Prior for Post-Training Quantization of Large Language Models
Date pendingArthur Negrão, Pedro Silva, Vander L. S. Freitas +2Transformer Architectures Post-Training
Postgraduate Program in Computer Science Federal University of Ouro Preto · Computing Department Federal University of Ouro Preto
AsyncFlow: An Asynchronous Streaming RL Framework for Efficient LLM Post-Training
Date pendingZhenyu Han, Ansheng You, Haibo Wang +16Asynchronous Execution Efficient Large Language Model
Huawei · Individual Researcher
TextAlign: Preference Alignment for Text Rendering with Hierarchical Rewards
Date pendingMingxuan Cui, Jingpu Yang, Fengxian Ji +7Image-Text Alignment Few-Shot Font Generation
Mohamed bin Zayed University of Artificial Intelligence · Chinese Academy of Sciences Institute of Automation · Northeastern University +1
Complementing reinforcement learning with SFT through logit averaging in the post training of LLMs
Date pendingXingwei Gan, Ying ZhuLogit Lens Post-Training
UC San Diego

May 9, 2026, cs.LGJ/K move · Enter open · S save

Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression

Hengyi Zhu, Zhendong Mi, Grace Li Zhang, Shaoyi Huang

1Stevens Institute of Technology, Hoboken, USA · 2Technical University of Darmstadt, Darmstadt, Germany

Large Language Model Compression Data Compression Methods Decode Speedup Post-Training

Open paper AskPDF arXiv

Large language models (LLMs) have rapidly grown in scale, creating substantial memory and computational costs that hinder efficient deployment. Singular value decomposition (SVD) has emerged as an effective post-training compression technique, but existing SVD-based methods rely on static rank truncation, applying a fixed prefix of singular components to all inputs regardless of their diversity. We identify two limitations of this static design: the optimal rank varies across individual prompts, and the selected rank is sensitive to the choice of calibration set, leading to suboptimal performance across diverse inputs. To address these challenges, we propose

\textbf{PARSE}

, a post-training framework for

\textbf{P}

rompt-

\textbf{A}

ware

\textbf{R}

ank

\textbf{S}

election as

\textbf{E}

xperts in SVD-compressed LLMs. PARSE trains a linear router offline to perform prompt-aware rank selection, decoupling it from calibration information by supervising the router against dense-model outputs on a large-scale corpus. We further observe that rank-selection patterns are shared across semantically similar prompts and remain stable across decoding steps, allowing appropriate rank subsets to be served directly from a pattern cache at inference. Complemented by expert memory aggregation and kernel fusion for system-level efficiency, PARSE is orthogonal to existing SVD-based pipelines and consistently improves both model quality and inference efficiency. Integrated with four representative SVD-based methods, PARSE improves average task accuracy by up to 10% at a compression ratio of 0.6 on LLaMA-7B, and achieves up to 2.5

\times

prefill and 2.4

\times

decode speedup over native SVD execution.

Momentum

Related topics

Latest papers 297