{"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/large-vision-language-models-are-unsupervised","title":"Large (Vision) Language Models are Unsupervised In-Context Learners","arxiv_id":"2504.02349","date":"2025-04-03","proceeding":null,"authors":["Artyom Gadetsky","Andrei Atanov","Yulun Jiang","Zhitong Gao","Ghazal Hosseini Mighan","Amir Zamir","Maria Brbic"],"abstract":"Recent advances in large language and vision-language models have enabled zero-shot inference, allowing models to solve new tasks without task-specific training. Various adaptation techniques such as prompt engineering, In-Context Learning (ICL), and supervised fine-tuning can further enhance the model's performance on a downstream task, but they require substantial manual effort to construct effective prompts or labeled examples. In this work, we introduce a joint inference framework for fully unsupervised adaptation, eliminating the need for manual prompt engineering and labeled examples. Unlike zero-shot inference, which makes independent predictions, the joint inference makes predictions simultaneously for all inputs in a given task. Since direct joint inference involves computationally expensive optimization, we develop efficient approximation techniques, leading to two unsupervised adaptation methods: unsupervised fine-tuning and unsupervised ICL. We demonstrate the effectiveness of our methods across diverse tasks and models, including language-only Llama-3.1 on natural language processing tasks, reasoning-oriented Qwen2.5-Math on grade school math problems, vision-language OpenFlamingo on vision tasks, and the API-only access GPT-4o model on massive multi-discipline tasks. Our experiments demonstrate substantial improvements over the standard zero-shot approach, including 39% absolute improvement on the challenging GSM8K math reasoning dataset. Remarkably, despite being fully unsupervised, our framework often performs on par with supervised approaches that rely on ground truth labels.","url_abs":"https://arxiv.org/abs/2504.02349v1","url_pdf":"https://arxiv.org/pdf/2504.02349v1.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":"large-vision-language-models-are-unsupervised","repo_url":"https://github.com/mlbio-epfl/joint-inference","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"gsm8k","task_name":"GSM8K"},{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"math","task_name":"Math"},{"task_slug":"prompt-engineering","task_name":"Prompt Engineering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2504.02349","atlas_url":"https://app.syntology.ai/?focus=2504.02349","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.02349"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2504.02349"},"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":"2504.02349"},"call":"get_harvested_code_for_paper(arxiv_id=\"2504.02349\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2504.02349"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2504.02349\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2504.02349"},"call":"get_concepts_for_paper(arxiv_id=\"2504.02349\")"}],"by_repo":[{"repo":"mlbio-epfl/joint-inference","kind":"official","samples":5,"ran":3,"instrument":2,"unverified":2}],"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","assemblies":[{"assembly_sha256":"e3cf660b6e2004be","source_kind":"generated","method_name":"unsupervised fine-tuning","realizes_method":"unsupervised fine-tuning","graph_method":"unsupervised fine-tuning","realizes_severed":false,"claim":"Compute the unsupervised ICL objective from Algorithm 1, line 7. Input log_probs_per_sample has shape (B, N, V) containing log probabilities log τ_θ(y | x_n^b) where B=batch size, N=samples per batch, V=vocabulary size. Input sampled_answers has shape (B, N) with integer indices in [0, V-1] represen","centrality":"contribution","entry":"compute_objective_with_prior_regularization","entry_sha256":"d207f13c1e22c0f9fe71bb190ce9bcd7b068ec00fca22204e3ea3631bdf098ed","n_bricks":1,"bricks":[{"entry":"compute_objective_with_prior_regularization","code_sha256":"d207f13c1e22c0f9fe71bb190ce9bcd7b068ec00fca22204e3ea3631bdf098ed","role":"entry","order":0,"level":1,"status":"ran"}],"ran_individually":1,"ran_individually_constructed":null,"ran_individually_by_status":{"ran":1},"end_to_end":{"ran":true,"level":1,"against_independent_peer":false,"n_peers":0,"v3_tests_passed_both_sides":0,"means":null},"level_lowered":{"level_was":2,"note":"lowered to the entry brick level: a single-routine assembly IS its brick","corrected_at":"2026-09-10T14:51:13.98Z"},"claim_match":"IMPLEMENTS","claim_match_rule":"single_judge","claim_match_basis":"construction","claim_match_judges":["construction:spec_drafted_for_this_method"],"label_verdict":"matches","label_verdict_judged":false,"suggested_method":null,"batch_at":"2026-09-07T20:44:12+00:00","sibling_group":"d5455e9cacf5","n_siblings":0,"retention":"keep","provenance":"assembly:generated_reference_impl_v1","attached_at":"2026-09-07T20:44:12+00:00","attached_provenance":"assembly:generated_reference_impl_v1"},{"assembly_sha256":"a00e03aebf19ad4b","source_kind":"harvested","method_name":"unsupervised ICL","realizes_method":"unsupervised ICL","graph_method":"unsupervised ICL","realizes_severed":false,"claim":"This code implements unsupervised in-context learning (ICL) by sampling examples from a training dataset and using them to infer labels for test queries without task-specific fine-tuning.","centrality":"contribution","entry":"icl_inference","entry_sha256":"09dd1dd500c5dec3","n_bricks":4,"bricks":[{"entry":"icl_inference","code_sha256":"09dd1dd500c5dec3","role":"entry","order":0,"level":0,"status":"unverified"},{"entry":"balanced_sampling","code_sha256":"83f821a080f6142a","role":"sampler","order":1,"level":0,"status":"unverified"},{"entry":"get_location_prefix","code_sha256":"640a420f34882a39","role":"helper","order":2,"level":1,"status":"ran_draft_wrong"},{"entry":"find_len_of_common_prefix","code_sha256":"cac6357ee909bfae","role":"helper","order":3,"level":1,"status":"ran_honours"}],"ran_individually":2,"ran_individually_constructed":0,"ran_individually_by_status":{"ran_draft_wrong":1,"ran_honours":1},"end_to_end":{"ran":false,"level":0},"level_lowered":null,"claim_match":"IMPLEMENTS","claim_match_rule":"single_judge","claim_match_basis":"judge:global.anthropic.claude-sonnet-4-5-20250929-v1:0","claim_match_judges":["global.anthropic.claude-sonnet-4-5-20250929-v1:0"],"label_verdict":"matches","label_verdict_judged":true,"suggested_method":null,"batch_at":"2026-09-09T22:44:46+00:00","sibling_group":"55f43ee67884","n_siblings":0,"retention":"keep","provenance":"assembly:harvest_method_v6","attached_at":"2026-09-09T22:44:46+00:00","attached_provenance":"assembly:harvest_method_v6"}],"assembly_sibling_groups":[{"group":"55f43ee67884","realizes_method":"unsupervised ICL","source_kind":"harvested","batch_at":"2026-09-09T22:44:46+00:00","members":["a00e03aebf19ad4b"]},{"group":"d5455e9cacf5","realizes_method":"unsupervised fine-tuning","source_kind":"generated","batch_at":"2026-09-07T20:44:12+00:00","members":["e3cf660b6e2004be"]}],"assemblies_is":{"about":"What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results.","harvested":"assembled from code harvested from a repository linked to the paper","generated":"generated from the paper text, not the repository's code","end_to_end":"for a harvested assembly, the assembly has not been run end-to-end; for a generated one, end_to_end.level is the assembly's own level and against_independent_peer says whether it ran against a peer: an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists","bricks_level":"each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level","ran_individually":"a COUNT of bricks that ran on their own on a fixture (split by status in ran_individually_by_status), not a level; instrument statuses are failures of Syntology's instrument, not of the code","siblings":"assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another","judged":"Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as \"by construction\" or \"not judged by a model\"","level_lowered":"a generated assembly whose level Syntology's graph lowered, with the level it had, the recorded reason and when","retention":"Syntology's spending triage for further verification, not a verdict","claim_match":"the claim_match vote: IMPLEMENTS, COMPONENT or UNCLEAR. Where claim_match_basis is \"judge:<model>\", it is one model's judgement of the assembly's label against the paper's claim (claim_match_rule single_judge, the model in claim_match_judges): not a verification and not a run. Where it is \"construction\", the value was set when the code was generated for this method and no model judged it, whatever claim_match_rule says. The page prints no claim match for any assembly","label_verdict":"whether the assembly's code is the method it is labelled with: matches, cannot_tell, mislabelled (suggested_method names what it looks like instead) or not_the_method; label_verdict_judged is true only when a model read the code (claim_match_basis \"judge:<model>\")","label_verdict_judged":"where true (claim_match_basis \"judge:<model>\"), label_verdict is a single model's judgement; not calibrated; not printed on the page; where false, it was set by construction, and the page prints it unless the node is shared with another paper (node_properties_written_for)","claim":"the claim text stored on the assembly node: for a harvested assembly, what the paper claims the code implements, as the lifting model wrote it; for a generated one, the specification the code was generated from. Not verified text","centrality":"how central the lifted method is to the paper, as the lifting step recorded it: contribution, component or incidental","v3_tests_passed_both_sides":"end_to_end.v3_tests_passed_both_sides: the number of property tests derived from the paper that passed on both implementations, as the attached report lists them; a level 3 card with 0 says its report lists none","ran_individually_constructed":"ran_individually_constructed: of the bricks counted in ran_individually, how many constructed an object rather than computing a result (Syntology's RAN record's rule, read from the graph-layer sample with the brick's code_sha256); a part of ran_individually, never subtracted from it and never a failure; null where a brick that ran has no sample in the graph layer this build read, so the figure was not counted"},"assemblies_call":{"tool":"assembly","arguments":{"arxiv_id":"2504.02349"},"call":"assembly(arxiv_id=\"2504.02349\")"},"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/2504.02349.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/2504.02349.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":"deterministic:regex_extraction","url":"https://github.com/mlbio-epfl/joint-inference","reach":null}],"summary":{"ran_draft_wrong":2,"ran_honours":1,"unverified":2},"n_samples":5,"n_ran":3,"n_constructed":0,"n_ran_checked":1,"n_instrument":2,"n_unverified":2,"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":5,"ran":3,"n_ran_checked":1,"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":0,"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":0,"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":2,"n_ran_without_level_date":1,"n_ran_signature_not_shown":0,"levels_written_between":{"first":"2026-09-22","last":"2026-09-22","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"},"samples":[{"code_sha256":"00a886c1703d3ae6","code_sha256_prefix":"00a886c1703d3ae6","entry":"replace_none_with_default","repo":"mlbio-epfl/joint-inference","repo_kind":"official","path":"unsupervised_ft_llm.py","file_url":"https://github.com/mlbio-epfl/joint-inference/blob/HEAD/unsupervised_ft_llm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"signature":"data:dict/0/?","fixture_out_type":"dict","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"00a886c1703d3ae6"}},{"code_sha256":"cac6357ee909bfae","code_sha256_prefix":"cac6357ee909bfae","entry":"find_len_of_common_prefix","repo":"mlbio-epfl/joint-inference","repo_kind":"official","path":"unsupervised_icl_llm.py","file_url":"https://github.com/mlbio-epfl/joint-inference/blob/HEAD/unsupervised_icl_llm.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"list1:list/1/int, list2:list/1/int","fixture_out_type":"int64","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"cac6357ee909bfae"}},{"code_sha256":"640a420f34882a39","code_sha256_prefix":"640a420f34882a39","entry":"get_location_prefix","repo":"mlbio-epfl/joint-inference","repo_kind":"official","path":"unsupervised_icl_llm.py","file_url":"https://github.com/mlbio-epfl/joint-inference/blob/HEAD/unsupervised_icl_llm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"template:callable/0/?, tokenizer:callable/0/?, input_dict:dict/0/?, y:str/0/str, prefix:str/0/str","fixture_out_type":"tuple","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"640a420f34882a39"}},{"code_sha256":"83f821a080f6142a","code_sha256_prefix":"83f821a080f6142a","entry":"balanced_sampling","repo":"mlbio-epfl/joint-inference","repo_kind":"official","path":"unsupervised_icl_llm.py","file_url":"https://github.com/mlbio-epfl/joint-inference/blob/HEAD/unsupervised_icl_llm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"83f821a080f6142a"}},{"code_sha256":"09dd1dd500c5dec3","code_sha256_prefix":"09dd1dd500c5dec3","entry":"icl_inference","repo":"mlbio-epfl/joint-inference","repo_kind":"official","path":"unsupervised_icl_llm.py","file_url":"https://github.com/mlbio-epfl/joint-inference/blob/HEAD/unsupervised_icl_llm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"09dd1dd500c5dec3"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"0135c703b786903bde81c147e5af46cc07c535be6eeac0fc51baeab9edacf90f","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}