{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/cls-rl-image-classification-with-rule-based","title":"Think or Not Think: A Study of Explicit Thinking in Rule-Based Visual Reinforcement Fine-Tuning","arxiv_id":"2503.16188","date":"2025-03-20","proceeding":null,"authors":["Ming Li","Jike Zhong","Shitian Zhao","Yuxiang Lai","Haoquan Zhang","Wang Bill Zhu","Kaipeng Zhang"],"abstract":"This paper investigates the role of explicit thinking process in rule-based reinforcement fine-tuning (RFT) for MLLMs. We first propose CLS-RL for MLLM image classification, using verifiable rewards for fine-tuning. Experiments show CLS-RL significantly outperforms SFT and yields a cross-dataset generalization effect. We then rethink and question whether explicit thinking in RFT is always necessary. Challenging the convention that explicit thinking is crucial for the success of RFT, we introduce No-Thinking-RL, exploring RFT without thinking by introducing a simple equality accuracy reward. We evaluate No-Thinking-RL on 6 diverse tasks across different model sizes and types. Experimental results reveal three key findings: 1). Visual perception tasks do not require thinking during RFT, as No-Thinking-RL consistently outperforms or matches Thinking-based RFT across model sizes. 2).} Models with limited capabilities struggle to generate high-quality CoT for RFT, making Thinking-based RFT less effective than No-Thinking-RL. 3). There are inconsistencies between the answers in the thinking and answer tags for some responses of thinking-based RFT, which show lower accuracy than the overall accuracy. We hypothesize that explicit thinking before verifiable answers may hinder reward convergence and reduce performance. To test this hypothesis, we propose Think-After-Answer, which places thinking after the answer to mitigate this effect for experimental verification. Lastly, we conduct a pilot study to explore whether MLLMs can learn when to think during RFT, introducing an Adaptive-Thinking method. Experiments show that it converges to a specific prompt depending on model capability and task complexity, achieving comparable or better performance than both Thinking and No-Thinking-RL. This suggests MLLMs can adaptively decide to think or not based on their capabilities and task complexity.","url_abs":"https://arxiv.org/abs/2503.16188v4","url_pdf":"https://arxiv.org/pdf/2503.16188v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"cls-rl-image-classification-with-rule-based","repo_url":"https://github.com/minglllli/CLS-RL","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"sft","method_name":"SFT"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.16188","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.16188"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/minglllli/CLS-RL","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":2,"ran_fixture":1},"by_repo_kind":{"official":{"samples":3,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":3,"samples":[{"code_sha256_prefix":"2bb2ff42bd737b97","entry":"extract_answer","repo":"minglllli/cls-rl","repo_kind":"official","path":"src/eval/test_cvbench.py","file_url":"https://github.com/minglllli/cls-rl/blob/HEAD/src/eval/test_cvbench.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2bb2ff42bd737b97"}},{"code_sha256_prefix":"674a8307721ab555","entry":"extract_characters_regex","repo":"minglllli/cls-rl","repo_kind":"official","path":"src/eval/test_cvbench.py","file_url":"https://github.com/minglllli/cls-rl/blob/HEAD/src/eval/test_cvbench.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"674a8307721ab555"}},{"code_sha256_prefix":"3f31b48fd862935d","entry":"get_image_data_url","repo":"minglllli/cls-rl","repo_kind":"official","path":"src/cls-rl/local_scripts/create_vision_cot_data.py","file_url":"https://github.com/minglllli/cls-rl/blob/HEAD/src/cls-rl/local_scripts/create_vision_cot_data.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3f31b48fd862935d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}