{"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/alignment-without-over-optimization-training","title":"Test-time Alignment of Diffusion Models without Reward Over-optimization","arxiv_id":"2501.05803","date":"2025-01-10","proceeding":null,"authors":["Sunwoo Kim","Minkyu Kim","Dongmin Park"],"abstract":"Diffusion models excel in generative tasks, but aligning them with specific objectives while maintaining their versatility remains challenging. Existing fine-tuning methods often suffer from reward over-optimization, while approximate guidance approaches fail to optimize target rewards effectively. Addressing these limitations, we propose a training-free, test-time method based on Sequential Monte Carlo (SMC) to sample from the reward-aligned target distribution. Our approach, tailored for diffusion sampling and incorporating tempering techniques, achieves comparable or superior target rewards to fine-tuning methods while preserving diversity and cross-reward generalization. We demonstrate its effectiveness in single-reward optimization, multi-objective scenarios, and online black-box optimization. This work offers a robust solution for aligning diffusion models with diverse downstream objectives without compromising their general capabilities. Code is available at https://github.com/krafton-ai/DAS.","url_abs":"https://arxiv.org/abs/2501.05803v3","url_pdf":"https://arxiv.org/pdf/2501.05803v3.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":"alignment-without-over-optimization-training","repo_url":"https://github.com/krafton-ai/das","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2501.05803","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2501.05803"}},"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":"deterministic:regex_extraction","url":"https://github.com/krafton-ai/DAS","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/krafton-ai/das","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":1,"unverified":11},"by_repo_kind":{"official":{"samples":12,"ran":1,"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":0,"samples":[{"code_sha256_prefix":"fde7881344a30b12","entry":"betas_for_alpha_bar","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/diffusers_patch/lcm_with_logprob.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/diffusers_patch/lcm_with_logprob.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fde7881344a30b12"}},{"code_sha256_prefix":"47714310f035f682","entry":"aesthetic_score","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/rewards.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/rewards.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"47714310f035f682"}},{"code_sha256_prefix":"b9541044b78a4a5c","entry":"clip_score","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/rewards.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/rewards.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b9541044b78a4a5c"}},{"code_sha256_prefix":"638ef43222c986d4","entry":"compute_ess","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/smc_utils.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/smc_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"638ef43222c986d4"}},{"code_sha256_prefix":"b7ecc0cc64dc5ebc","entry":"compute_ess_from_log_w","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/smc_utils.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/smc_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b7ecc0cc64dc5ebc"}},{"code_sha256_prefix":"986e249fed0367ec","entry":"ddim_prediction_with_mean","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/diffusers_patch/ddim_with_logprob.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/diffusers_patch/ddim_with_logprob.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"986e249fed0367ec"}},{"code_sha256_prefix":"d3b6a569179d9722","entry":"ddim_step_with_mean","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/diffusers_patch/ddim_with_logprob.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/diffusers_patch/ddim_with_logprob.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d3b6a569179d9722"}},{"code_sha256_prefix":"0f88113ed6486c35","entry":"get_variance","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/diffusers_patch/ddim_with_logprob.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/diffusers_patch/ddim_with_logprob.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0f88113ed6486c35"}},{"code_sha256_prefix":"e1a14b1c366e53ca","entry":"jpeg_compressibility","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/rewards.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/rewards.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e1a14b1c366e53ca"}},{"code_sha256_prefix":"1a8d52ace26b455e","entry":"lcm_step_with_mean","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/diffusers_patch/lcm_with_logprob.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/diffusers_patch/lcm_with_logprob.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1a8d52ace26b455e"}},{"code_sha256_prefix":"c05fb5231070387d","entry":"normalize_weights","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/smc_utils.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/smc_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c05fb5231070387d"}},{"code_sha256_prefix":"6cef65ea9b3a0d85","entry":"rescale_zero_terminal_snr","repo":"krafton-ai/DAS","repo_kind":"official","path":"das/diffusers_patch/lcm_with_logprob.py","file_url":"https://github.com/krafton-ai/DAS/blob/HEAD/das/diffusers_patch/lcm_with_logprob.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6cef65ea9b3a0d85"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}