{"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/arxiv-2510-22860","title":"Far from the Shallow: Brain-Predictive Reasoning Embedding through Residual Disentanglement","arxiv_id":"2510.22860","date":"2025-10-26","proceeding":"NeurIPS","authors":["Linyang He","Tianjun Zhong","Richard Antonello","Gavin Mischler","Micah Goldblum","Nima Mesgarani"],"abstract":"Understanding how the human brain progresses from processing simple linguistic inputs to performing high-level reasoning is a fundamental challenge in neuroscience. While modern large language models (LLMs) are increasingly used to model neural responses to language, their internal representations are highly \"entangled,\" mixing information about lexicon, syntax, meaning, and reasoning. This entanglement biases conventional brain encoding analyses toward linguistically shallow features (e.g., lexicon and syntax), making it difficult to isolate the neural substrates of cognitively deeper processes. Here, we introduce a residual disentanglement method that computationally isolates these components. By first probing an LM to identify feature-specific layers, our method iteratively regresses out lower-level representations to produce four nearly orthogonal embeddings for lexicon, syntax, meaning, and, critically, reasoning. We used these disentangled embeddings to model intracranial (ECoG) brain recordings from neurosurgical patients listening to natural speech. We show that: 1) This isolated reasoning embedding exhibits unique predictive power, accounting for variance in neural activity not explained by other linguistic features and even extending to the recruitment of visual regions beyond classical language areas. 2) The neural signature for reasoning is temporally distinct, peaking later (~350-400ms) than signals related to lexicon, syntax, and meaning, consistent with its position atop a processing hierarchy. 3) Standard, non-disentangled LLM embeddings can be misleading, as their predictive success is primarily attributable to linguistically shallow features, masking the more subtle contributions of deeper cognitive processing.","url_abs":"https://arxiv.org/abs/2510.22860","url_pdf":"https://arxiv.org/pdf/2510.22860","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2510.22860","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2510.22860"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/alexwarstadt/blimp","reach":{"status":"ok"}},{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/kanishkamisra/comps","reach":null},{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/calclavia/tal-asrd","reach":{"status":"ok"}}],"summary":{"unverified":11},"by_repo_kind":{"found_in_text":{"samples":11,"ran":0,"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":11,"samples":[{"code_sha256_prefix":"639778f08433f3d0","entry":"compute_sequence_match","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/wder.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/wder.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"639778f08433f3d0"}},{"code_sha256_prefix":"902aafeef001368f","entry":"freq_mask","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/asr/models.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/asr/models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"902aafeef001368f"}},{"code_sha256_prefix":"b26dcac6897f2d7f","entry":"get_argparser","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/asr/args.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/asr/args.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b26dcac6897f2d7f"}},{"code_sha256_prefix":"1d874f65af11de22","entry":"get_list_inverse_index","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/wder.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/wder.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1d874f65af11de22"}},{"code_sha256_prefix":"f05313b6b339e5ba","entry":"inv_sqrt_schedule","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/schedules.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/schedules.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f05313b6b339e5ba"}},{"code_sha256_prefix":"977dbb4b29bf2bf6","entry":"pairwise_idp","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/wder_search.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/wder_search.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"977dbb4b29bf2bf6"}},{"code_sha256_prefix":"3570c159afeb008f","entry":"pairwise_ndp","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/wder_search.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/wder_search.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3570c159afeb008f"}},{"code_sha256_prefix":"d43d9fab1ea04f5d","entry":"time_mask","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/asr/models.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/asr/models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d43d9fab1ea04f5d"}},{"code_sha256_prefix":"d0e80b855a08c984","entry":"tokenize","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/alignment/aeneas.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/alignment/aeneas.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d0e80b855a08c984"}},{"code_sha256_prefix":"8e7b6595aa28158d","entry":"triangle_schedule","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/schedules.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/schedules.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8e7b6595aa28158d"}},{"code_sha256_prefix":"441d2156007039fa","entry":"tweet_tokenize","repo":"calclavia/tal-asrd","repo_kind":"found_in_text","path":"tal/wder.py","file_url":"https://github.com/calclavia/tal-asrd/blob/HEAD/tal/wder.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"441d2156007039fa"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CL","source":"arxiv_api"},"syntology_extracted_results":null}