# Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training

arXiv: https://arxiv.org/abs/2507.05386 · HTML: https://mlsanity.com/paper/2507.05386/

## Key facts

- Cited by 47 papers in the map (data through 14 Oct 2026).
- Builds on 48 of its 68 references that are in the map.
- First posted on arXiv 7 Jul 2025.

## Builds on (most cited first, 30 of 48)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark](https://mlsanity.com/paper/2406.01574/) (2024)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [Understanding R1-Zero-Like Training: A Critical Perspective](https://mlsanity.com/paper/2503.20783/) (2025)
- [Prioritized Experience Replay](https://mlsanity.com/paper/1511.05952/) (2015)
- [VizWiz Grand Challenge: Answering Visual Questions from Blind People](https://mlsanity.com/paper/1802.08218/) (2018)
- [SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training](https://mlsanity.com/paper/2501.17161/) (2025)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [PathVQA: 30000+ Questions for Medical Visual Question Answering](https://mlsanity.com/paper/2003.10286/) (2020)
- [Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning](https://mlsanity.com/paper/2105.04165/) (2021)

## Built on by (most cited first, 30 of 47)

- [Self-Distillation Enables Continual Learning](https://mlsanity.com/paper/2601.19897/) (2026)
- [RL's Razor: Why Online Reinforcement Learning Forgets Less](https://mlsanity.com/paper/2509.04259/) (2025)
- [Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting](https://mlsanity.com/paper/2510.18874/) (2025)
- [LLMs are Single-threaded Reasoners: Demystifying the Working Mechanism of Soft Thinking](https://mlsanity.com/paper/2508.03440/) (2025)
- [RL Squeezes, SFT Expands: A Comparative Study of Reasoning LLMs](https://mlsanity.com/paper/2509.21128/) (2025)
- [Continual Learning via Sparse Memory Finetuning](https://mlsanity.com/paper/2510.15103/) (2025)
- [Simple Recipe Works: Vision-Language-Action Models are Natural Continual Learners with Reinforcement Learning](https://mlsanity.com/paper/2603.11653/) (2026)
- [Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability](https://mlsanity.com/paper/2604.06628/) (2026)
- [Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2602.10503/) (2026)
- [VLA-OPD: Bridging Offline SFT and Online RL for Vision-Language-Action Models via On-Policy Distillation](https://mlsanity.com/paper/2603.26666/) (2026)
- [Reinforcement Learning on Pre-Training Data](https://mlsanity.com/paper/2509.19249/) (2025)
- [RubiCap: Rubric-Guided Reinforcement Learning for Dense Image Captioning](https://mlsanity.com/paper/2603.09160/) (2026)
- [The Finetuner's Fallacy: When to Pretrain with Your Finetuning Data](https://mlsanity.com/paper/2603.16177/) (2026)
- [Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs](https://mlsanity.com/paper/2509.23657/) (2025)
- [Zoom-Zero: Reinforced Coarse-to-Fine Video Understanding via Temporal Zoom-in](https://mlsanity.com/paper/2512.14273/) (2025)
- [Learning with Rare Success but Rich Feedback via Reflection-Enhanced Self-Distillation](https://mlsanity.com/paper/2605.12741/) (2026)
- [Continual GUI Agents](https://mlsanity.com/paper/2601.20732/) (2026)
- [Denser neq Better: Limits of On-Policy Self-Distillation for Continual Post-Training](https://mlsanity.com/paper/2607.01763/) (2026)
- [InfoSFT: Learn More and Forget Less with Information-Aware Token Weighting](https://mlsanity.com/paper/2605.14967/) (2026)
- [Understanding Post-Training Structural Changes in Large Language Models](https://mlsanity.com/paper/2509.17866/) (2025)
- [Learning to Refuse: Refusal-Aware Reinforcement Fine-Tuning for Hard-Irrelevant Queries in Video Temporal Grounding](https://mlsanity.com/paper/2511.23151/) (2025)
- [CGL: Advancing Continual GUI Learning via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2603.02951/) (2026)
- [Rotation-Preserving Supervised Fine-Tuning](https://mlsanity.com/paper/2605.10973/) (2026)
- [Overcoming Forgetting in LLM Fine-Tuning with Evolution Strategies](https://mlsanity.com/paper/2605.30148/) (2026)
- [A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL](https://mlsanity.com/paper/2606.02398/) (2026)
- [Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning](https://mlsanity.com/paper/2605.09640/) (2026)
- [RL Forgets! Towards Continual Policy Optimization](https://mlsanity.com/paper/2607.04364/) (2026)
- [Towards Human-like Physical Intelligence: Lifelong Vision-Language-Action Learning for Robotic Manipulation](https://mlsanity.com/paper/2607.14852/) (2026)
- [SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs](https://mlsanity.com/paper/2608.03573/) (2026)
- [TMS: Trajectory-Mixed Supervision for Reward-Free, On-Policy SFT](https://mlsanity.com/paper/2602.03073/) (2026)

## Abstract (arXiv)

Continual post-training (CPT) is a popular and effective technique for adapting foundation models like multimodal large language models to ever-evolving downstream tasks. While existing research primarily focuses on methods like data replay, model expansion, or parameter regularization, the fundamental role of the learning paradigm remains largely unexplored. This paper presents a comparative analysis of two core post-training paradigms: supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT), investigating their respective impacts on knowledge retention during CPT. Our experiments are conducted across multiple multimodal tasks, utilizing Qwen2.5-VL-7B-Instruct as the base model. The investigation yields two significant findings: (1) When continuously learning on downstream tasks, SFT leads to catastrophic forgetting of previously learned tasks. In contrast, RFT inherently preserves prior knowledge and achieves performance comparable to multi-task training. (2) RFT successfully protects and even enhances the model's general knowledge on standard benchmarks, while SFT degrades general model capabilities severely. Further analysis reveals that this stability is not primarily due to explicit mechanisms like KL penalty or chain-of-thought reasoning. We investigate RFT's learning dynamics and find that its selective update mechanism inherently prevents interference with established knowledge. Based on this insight, we propose a rollout-based instance filtering algorithm (RIF-RFT) that enhances the training efficiency of RFT by focusing on learnable samples. Our comprehensive study demonstrates the superiority of RFT as a robust paradigm for continual post-training.