{"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/complexity-control-facilitates-reasoning","title":"Complexity Control Facilitates Reasoning-Based Compositional Generalization in Transformers","arxiv_id":"2501.08537","date":"2025-01-15","proceeding":null,"authors":["Zhongwang Zhang","Pengxiao Lin","Zhiwei Wang","Yaoyu Zhang","Zhi-Qin John Xu"],"abstract":"Transformers have demonstrated impressive capabilities across various tasks, yet their performance on compositional problems remains a subject of debate. In this study, we investigate the internal mechanisms underlying Transformers' behavior in compositional tasks. We find that complexity control strategies significantly influence whether the model learns primitive-level rules that generalize out-of-distribution (reasoning-based solutions) or relies solely on memorized mappings (memory-based solutions). By applying masking strategies to the model's information circuits and employing multiple complexity metrics, we reveal distinct internal working mechanisms associated with different solution types. Further analysis reveals that reasoning-based solutions exhibit a lower complexity bias, which aligns with the well-studied neuron condensation phenomenon. This lower complexity bias is hypothesized to be the key factor enabling these solutions to learn reasoning rules. We validate these conclusions across multiple real-world datasets, including image generation and natural language processing tasks, confirming the broad applicability of our findings.","url_abs":"https://arxiv.org/abs/2501.08537v1","url_pdf":"https://arxiv.org/pdf/2501.08537v1.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":"complexity-control-facilitates-reasoning","repo_url":"https://github.com/sjtuzzw/complexity_control","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2501.08537","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2501.08537"}},"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/sjtuzzw/complexity_control","reach":null}],"summary":{"ran_draft_wrong":2,"ran_violates":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":"e072a18b589baed4","entry":"attn_mask","repo":"sjtuzzw/complexity_control","repo_kind":"official","path":"code_for_anchor_func/model/GPT2_init_for_diff_part_prenorm.py","file_url":"https://github.com/sjtuzzw/complexity_control/blob/HEAD/code_for_anchor_func/model/GPT2_init_for_diff_part_prenorm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e072a18b589baed4"}},{"code_sha256_prefix":"e6aecf6de669e5bf","entry":"get_attn_pad_mask","repo":"sjtuzzw/complexity_control","repo_kind":"official","path":"code_for_anchor_func/model/GPT2_init_for_diff_part_prenorm.py","file_url":"https://github.com/sjtuzzw/complexity_control/blob/HEAD/code_for_anchor_func/model/GPT2_init_for_diff_part_prenorm.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e6aecf6de669e5bf"}},{"code_sha256_prefix":"feef287e9f210cfe","entry":"get_attn_subsequence_mask","repo":"sjtuzzw/complexity_control","repo_kind":"official","path":"code_for_anchor_func/model/GPT2_init_for_diff_part_prenorm.py","file_url":"https://github.com/sjtuzzw/complexity_control/blob/HEAD/code_for_anchor_func/model/GPT2_init_for_diff_part_prenorm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"feef287e9f210cfe"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}