{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/scaling-up-test-time-compute-with-latent","title":"Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach","arxiv_id":"2502.05171","date":"2025-02-07","proceeding":null,"authors":["Jonas Geiping","Sean McLeish","Neel Jain","John Kirchenbauer","Siddharth Singh","Brian R. Bartoldson","Bhavya Kailkhura","Abhinav Bhatele","Tom Goldstein"],"abstract":"We study a novel language model architecture that is capable of scaling test-time computation by implicitly reasoning in latent space. Our model works by iterating a recurrent block, thereby unrolling to arbitrary depth at test-time. This stands in contrast to mainstream reasoning models that scale up compute by producing more tokens. Unlike approaches based on chain-of-thought, our approach does not require any specialized training data, can work with small context windows, and can capture types of reasoning that are not easily represented in words. We scale a proof-of-concept model to 3.5 billion parameters and 800 billion tokens. We show that the resulting model can improve its performance on reasoning benchmarks, sometimes dramatically, up to a computation load equivalent to 50 billion parameters.","url_abs":"https://arxiv.org/abs/2502.05171v1","url_pdf":"https://arxiv.org/pdf/2502.05171v1.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":"scaling-up-test-time-compute-with-latent","repo_url":"https://github.com/seal-rg/recurrent-pretraining","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"scaling-up-test-time-compute-with-latent","repo_url":"https://github.com/gair-nlp/prox","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2502.05171","atlas_url":"https://app.syntology.ai/?focus=2502.05171","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2502.05171"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2502.05171"},"returns":"the samples harvested from this paper, each with its run record; it does not verify the paper"},{"tool":"get_code","arguments_in":"samples[].code_sha256","returns":"one stored sample by its code_sha256; not shown on this site"}]},"developers":"https://syntology.ai/developers","held":true,"reason":"in_graph_layer","checked_against":null,"agent_calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2502.05171"},"call":"get_harvested_code_for_paper(arxiv_id=\"2502.05171\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2502.05171"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2502.05171\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2502.05171"},"call":"get_concepts_for_paper(arxiv_id=\"2502.05171\")"}],"by_repo":[{"repo":"seal-rg/recurrent-pretraining","kind":"official","samples":13,"ran":10,"instrument":2,"unverified":3},{"repo":"gair-nlp/prox","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0,"harvested_for_other_papers":1,"harvested_for":["2504.11393"],"harvested_for_calls":[{"arxiv_id":"2504.11393","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2504.11393"},"call":"get_harvested_code_for_paper(arxiv_id=\"2504.11393\")"}]}],"by_repo_harvested_for_is":"harvested_for: the arXiv ids (as Syntology's graph holds them) of every paper whose samples Syntology's graph links to this repository, in arXiv-id order, complete; the page names the first 3 and says how many more. A link only: it does not say the papers are the same work, that one copied the other, or that the code implements this paper's method; harvested_for_calls gives the get_harvested_code_for_paper call for each arXiv-shaped id among them","by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code","read_at":"2026-10-05T00:36:11+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","ran_record":{"url":"https://syntology.ai/api/ran/2502.05171.json","schema":"syntology.ran_record/1","counts_crosswalk":{"n_samples":"syntology.counts.lifted","n_ran":"syntology.counts.executed","n_ran_checked":"syntology.counts.checked","n_instrument":"syntology.counts.instrument_failures","n_unverified":"syntology.counts.no_recorded_run"},"note":"live record, measured at request time; may be newer than this page's graph read","badge":"https://syntology.ai/api/ran/2502.05171.svg"},"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/gair-nlp/prox","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/seal-rg/recurrent-pretraining","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":8,"ran_draft_wrong":2,"unverified":3},"n_samples":13,"n_ran":10,"n_constructed":0,"n_ran_checked":8,"n_instrument":2,"n_unverified":3,"n_instrument_is":"failures of Syntology's instrument, not of the code","every_run_is_an_instrument_failure":false,"key_notes":{"n_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","n_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"by_repo_kind":{"official":{"samples":13,"ran":10,"n_ran_checked":8,"n_instrument":2,"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":4,"inline_ok_is":"inline_ok is true only when both licence gates allow Syntology to serve this code's text: this paper's link (inline_ok_link, from the licence of this paper's copy) and the stored copy of the same code that Syntology serves, which may have been first harvested from another repository. withheld_clause is Syntology's own wording of which gate keeps the text a pointer and why, and is null when inline_ok is true. This site shows no code text either way; n_pointer_only_for_licence counts samples with inline_ok false (both gates), n_pointer_only_link_allows those of them whose inline_ok_link is true","n_pointer_only_link_allows":4,"fixture_fields_is":{"signature_and_fixture_out_type":"the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right","level_recorded_at":"when Syntology's graph wrote the sample's verification level, not when the code executed; absent where the level was written before the date was kept, and never filled in with the build date","present_on":"samples that ran only; a key is absent, never empty, where the graph records nothing"},"n_ran_with_level_date":10,"n_ran_without_level_date":0,"n_ran_signature_not_shown":0,"levels_written_between":{"first":"2026-09-22","last":"2026-09-25","is":"the first and last UTC dates on which Syntology's graph wrote a verification level for a sample on this page that ran, not when the code executed"},"code_page_min_papers":5,"code_page_is":"a page for the entry name across every paper's harvested code; the name is a string, not an identity","samples":[{"code_sha256":"d5b72c13e150ef82","code_sha256_prefix":"d5b72c13e150ef82","entry":"apply_rotary_emb_complex_like","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"recpre/legacy_modeling_file.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/recpre/legacy_modeling_file.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"q:Tensor/3/float, k:Tensor/3/float, freqs_cis:Tensor/4/float","fixture_out_type":"tuple","level_recorded_at":"2026-09-25T16:13:10.526492+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from wenquanlu/huginn-latent-cot, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"d5b72c13e150ef82"}},{"code_sha256":"3e3df5fbf65b64b2","code_sha256_prefix":"3e3df5fbf65b64b2","entry":"frontier_max_minutes","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"launch_frontier.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/launch_frontier.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"num_nodes:int/0/int","fixture_out_type":"int","level_recorded_at":"2026-09-25T04:30:57.282821+00:00","retrieve_with":{"tool":"get_code","code_sha256":"3e3df5fbf65b64b2"},"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"3e3df5fbf65b64b2"}},{"code_sha256":"2577d131ae9699ea","code_sha256_prefix":"2577d131ae9699ea","entry":"get_comms_and_slingshot","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"launch_frontier.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/launch_frontier.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"installdir:str/0/str, enable_net_gdr:bool/0/bool, debug_flags:bool/0/bool, rccl_algo:NoneType/0/?","fixture_out_type":"str","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"2577d131ae9699ea"}},{"code_sha256":"ada420946a965abb","code_sha256_prefix":"ada420946a965abb","entry":"get_unwrapped_model","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"finetuning_simple_example.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/finetuning_simple_example.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"signature":"state:dict/0/?","fixture_out_type":"SimpleModel","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from njuyxw/STARS, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"ada420946a965abb"}},{"code_sha256":"00abb7bc07b05163","code_sha256_prefix":"00abb7bc07b05163","entry":"load_specific_keys","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"evaluate_raven/quick_chat_rec_compare.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/evaluate_raven/quick_chat_rec_compare.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"checkpoint_name:str/0/str, keys:list/1/str","fixture_out_type":"dict","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"00abb7bc07b05163"}},{"code_sha256":"7b9bbef7f2314381","code_sha256_prefix":"7b9bbef7f2314381","entry":"load_standard_modules","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"launch_frontier.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/launch_frontier.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"rocm_version:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","retrieve_with":{"tool":"get_code","code_sha256":"7b9bbef7f2314381"},"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"7b9bbef7f2314381"}},{"code_sha256":"400ea4f64ccce5ef","code_sha256_prefix":"400ea4f64ccce5ef","entry":"parse_eval_args","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"evaluate_raven/local_lm_eval.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/evaluate_raven/local_lm_eval.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"signature":"parser:ArgumentParser/0/?","fixture_out_type":"Namespace","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","code_page":"/code/parse-eval-args","census":null,"licence":"Apache-2.0","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from BatsResearch/crosslingual-test-time-scaling, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"400ea4f64ccce5ef"}},{"code_sha256":"a4a69f170428494a","code_sha256_prefix":"a4a69f170428494a","entry":"parse_gen_kwargs","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"evaluate_raven/record_steps_in_bench.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/evaluate_raven/record_steps_in_bench.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"gen_kwargs_str:str/0/str","fixture_out_type":"dict","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","retrieve_with":{"tool":"get_code","code_sha256":"a4a69f170428494a"},"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"a4a69f170428494a"}},{"code_sha256":"db336122f09999c7","code_sha256_prefix":"db336122f09999c7","entry":"precompute_freqs_cis","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"recpre/legacy_modeling_file.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/recpre/legacy_modeling_file.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"dim:int/0/int, end:int/0/int, theta:float/0/float, condense_ratio:int/0/int","fixture_out_type":"Tensor","level_recorded_at":"2026-09-25T16:13:10.526492+00:00","code_page":"/code/precompute-freqs-cis","census":null,"licence":"Apache-2.0","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from wenquanlu/huginn-latent-cot, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"db336122f09999c7"}},{"code_sha256":"40881a9c742670e8","code_sha256_prefix":"40881a9c742670e8","entry":"sample_gamma","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"recpre/model_dynamic.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/recpre/model_dynamic.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"alpha:float/0/float, beta:float/0/float","fixture_out_type":"Tensor","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"40881a9c742670e8"}},{"code_sha256":"7f7dafd2517a1906","code_sha256_prefix":"7f7dafd2517a1906","entry":"get_adaptive_exit_evaluator","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"recpre/raven_modeling_minimal.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/recpre/raven_modeling_minimal.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"7f7dafd2517a1906"}},{"code_sha256":"78396767cd633b79","code_sha256_prefix":"78396767cd633b79","entry":"startup","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"finetuning_simple_example.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/finetuning_simple_example.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":"/code/startup","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"78396767cd633b79"}},{"code_sha256":"06b670bf334b8d88","code_sha256_prefix":"06b670bf334b8d88","entry":"try_parse_json","repo":"seal-rg/recurrent-pretraining","repo_kind":"official","path":"evaluate_raven/local_lm_eval.py","file_url":"https://github.com/seal-rg/recurrent-pretraining/blob/HEAD/evaluate_raven/local_lm_eval.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":"/code/try-parse-json","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"06b670bf334b8d88"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"6fc9ffe0ab857da4eee43062dd65e13b63edff722b1e71768f812c16b3639707","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}