{"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":"/task/hellaswag","name":"HellaSwag","slug":"hellaswag","description_markdown":null,"categories":[{"name":"Natural Language Processing","url":"/area/natural-language-processing"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"derived"},"counts":{"papers_tagged":39,"papers_with_code":21,"benchmarks":0,"benchmark_tables_in_archive":0,"benchmark_tables_shown":0,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":0,"subtasks":0,"parent_tasks":1},"benchmarks":[],"datasets":[],"subtasks":[],"parent_tasks":[{"url":"/task/common-sense-reasoning","name":"Common Sense Reasoning"}],"papers":{"order":"repositories listed in the archive (most first), then date (newest first), then slug; the archive holds no stars","population":"every paper the archive tags with this task (distinct papers; spam titles not listed). 'of' counts the papers listed; until 2026-09-25 it counted only the papers with a code link, and that number is of_with_code","page":1,"pages_in_order":1,"rows_per_page":100,"shown":39,"of":39,"of_with_code":21,"tagged_in_all":39,"counts":{"archive_papers_tagged":39,"with_a_code_link":21,"where_syntology_ran_a_sample":8,"not_listed_spam_title":0,"listed":39,"listed_where_code_ran":8,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":7,"every_run_a_failure_of_syntologys_instrument":1,"listed_with_a_run_with_no_instrument_failure":7,"listed_every_run_a_failure_of_syntologys_instrument":1,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers the archive tags; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"items":[{"url":"/paper/scaling-language-models-methods-analysis-1","slug":"scaling-language-models-methods-analysis-1","title":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","date":"2021-12-08","arxiv_id":"2112.11446","repositories_listed":3,"syntology":null},{"url":"/paper/training-compute-optimal-large-language","slug":"training-compute-optimal-large-language","title":"Training Compute-Optimal Large Language Models","date":"2022-03-29","arxiv_id":"2203.15556","repositories_listed":2,"syntology":{"n":11,"n_ran":8,"n_constructed":3,"n_ran_checked":5,"n_instrument":3,"n_unverified":3,"n_honours":2,"n_violates":0,"n_no_contract":3,"n_pointer_only":4,"phrase":"8 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified","sample_list":"/paper/training-compute-optimal-large-language#ran","syntology_url":"https://syntology.ai/paper/2203.15556","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.15556"}},"official":null}},{"url":"/paper/hellaswag-can-a-machine-really-finish-your","slug":"hellaswag-can-a-machine-really-finish-your","title":"HellaSwag: Can a Machine Really Finish Your Sentence?","date":"2019-05-19","arxiv_id":"1905.07830","repositories_listed":2,"syntology":{"n":6,"n_ran":4,"n_constructed":0,"n_ran_checked":4,"n_instrument":0,"n_unverified":2,"n_honours":0,"n_violates":0,"n_no_contract":4,"n_pointer_only":4,"phrase":"4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified","sample_list":"/paper/hellaswag-can-a-machine-really-finish-your#ran","syntology_url":"https://syntology.ai/paper/1905.07830","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1905.07830"}},"official":null}},{"url":"/paper/simulating-training-data-leakage-in-multiple","slug":"simulating-training-data-leakage-in-multiple","title":"Simulating Training Data Leakage in Multiple-Choice Benchmarks for LLM Evaluation","date":"2025-05-30","arxiv_id":"2505.24263","repositories_listed":1,"syntology":null},{"url":"/paper/datadecide-how-to-predict-best-pretraining","slug":"datadecide-how-to-predict-best-pretraining","title":"DataDecide: How to Predict Best Pretraining Data with Small Experiments","date":"2025-04-15","arxiv_id":"2504.11393","repositories_listed":1,"syntology":{"n":14,"n_ran":11,"n_constructed":0,"n_ran_checked":10,"n_instrument":1,"n_unverified":3,"n_honours":1,"n_violates":1,"n_no_contract":8,"n_pointer_only":1,"phrase":"11 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified","sample_list":"/paper/datadecide-how-to-predict-best-pretraining#ran","syntology_url":"https://syntology.ai/paper/2504.11393","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.11393"}},"official":null}},{"url":"/paper/what-the-hellaswag-on-the-validity-of-common","slug":"what-the-hellaswag-on-the-validity-of-common","title":"What the HellaSwag? On the Validity of Common-Sense Reasoning Benchmarks","date":"2025-04-10","arxiv_id":"2504.07825","repositories_listed":1,"syntology":null},{"url":"/paper/finerweb-10bt-refining-web-data-with-llm","slug":"finerweb-10bt-refining-web-data-with-llm","title":"FinerWeb-10BT: Refining Web Data with LLM-Based Line-Level Filtering","date":"2025-01-13","arxiv_id":"2501.07314","repositories_listed":1,"syntology":null},{"url":"/paper/lora-done-rite-robust-invariant","slug":"lora-done-rite-robust-invariant","title":"LoRA Done RITE: Robust Invariant Transformation Equilibration for LoRA Optimization","date":"2024-10-27","arxiv_id":"2410.20625","repositories_listed":1,"syntology":{"n":3,"n_ran":2,"n_constructed":1,"n_ran_checked":1,"n_instrument":1,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":1,"n_pointer_only":3,"phrase":"2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/lora-done-rite-robust-invariant#ran","syntology_url":"https://syntology.ai/paper/2410.20625","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.20625"}},"official":{"repos":["gkevinyen5418/LoRA-RITE"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":1,"n_ran_no_instrument_failure":1,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/you-can-remove-gpt2-s-layernorm-by-fine","slug":"you-can-remove-gpt2-s-layernorm-by-fine","title":"You can remove GPT2's LayerNorm by fine-tuning","date":"2024-09-06","arxiv_id":"2409.13710","repositories_listed":1,"syntology":null},{"url":"/paper/texttt-metabench-a-sparse-benchmark-to","slug":"texttt-metabench-a-sparse-benchmark-to","title":"$\\texttt{metabench}$ -- A Sparse Benchmark to Measure General Ability in Large Language Models","date":"2024-07-04","arxiv_id":"2407.12844","repositories_listed":1,"syntology":{"n":9,"n_ran":8,"n_constructed":0,"n_ran_checked":3,"n_instrument":5,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":3,"n_pointer_only":9,"phrase":"8 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/texttt-metabench-a-sparse-benchmark-to#ran","syntology_url":"https://syntology.ai/paper/2407.12844","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.12844"}},"official":{"repos":["adkipnis/metabench"],"state":"official (archive's flag): 8 ran","n_ran":8,"n_constructed":0,"n_ran_no_instrument_failure":3,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/layer-skip-enabling-early-exit-inference-and","slug":"layer-skip-enabling-early-exit-inference-and","title":"LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding","date":"2024-04-25","arxiv_id":"2404.16710","repositories_listed":1,"syntology":{"n":5,"n_ran":4,"n_constructed":1,"n_ran_checked":1,"n_instrument":3,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":1,"n_pointer_only":5,"phrase":"4 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/layer-skip-enabling-early-exit-inference-and#ran","syntology_url":"https://syntology.ai/paper/2404.16710","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2404.16710"}},"official":{"repos":["facebookresearch/layerskip"],"state":"official (archive's flag): 4 ran","n_ran":4,"n_constructed":1,"n_ran_no_instrument_failure":1,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/sage-evaluating-moral-consistency-in-large","slug":"sage-evaluating-moral-consistency-in-large","title":"SaGE: Evaluating Moral Consistency in Large Language Models","date":"2024-02-21","arxiv_id":"2402.13709","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_constructed":0,"n_ran_checked":0,"n_instrument":1,"n_unverified":0,"n_honours":0,"n_violates":0,"n_no_contract":0,"n_pointer_only":1,"phrase":"1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified","sample_list":"/paper/sage-evaluating-moral-consistency-in-large#ran","syntology_url":"https://syntology.ai/paper/2402.13709","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.13709"}},"official":{"repos":["vnnm404/SaGE"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["official"]}}},{"url":"/paper/attacks-on-node-attributes-in-graph-neural","slug":"attacks-on-node-attributes-in-graph-neural","title":"Attacks on Node Attributes in Graph Neural Networks","date":"2024-02-19","arxiv_id":"2402.12426","repositories_listed":1,"syntology":null},{"url":"/paper/gemini-in-reasoning-unveiling-commonsense-in","slug":"gemini-in-reasoning-unveiling-commonsense-in","title":"Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Models","date":"2023-12-29","arxiv_id":"2312.17661","repositories_listed":1,"syntology":null},{"url":"/paper/an-open-source-data-contamination-report-for","slug":"an-open-source-data-contamination-report-for","title":"An Open Source Data Contamination Report for Large Language Models","date":"2023-10-26","arxiv_id":"2310.17589","repositories_listed":1,"syntology":null},{"url":"/paper/in-contextual-bias-suppression-for-large","slug":"in-contextual-bias-suppression-for-large","title":"In-Contextual Gender Bias Suppression for Large Language Models","date":"2023-09-13","arxiv_id":"2309.07251","repositories_listed":1,"syntology":null},{"url":"/paper/toward-adversarial-training-on-contextualized","slug":"toward-adversarial-training-on-contextualized","title":"Toward Adversarial Training on Contextualized Language Representation","date":"2023-05-08","arxiv_id":"2305.04557","repositories_listed":1,"syntology":null},{"url":"/paper/grada-graph-generative-data-augmentation-for","slug":"grada-graph-generative-data-augmentation-for","title":"GraDA: Graph Generative Data Augmentation for Commonsense Reasoning","date":"2022-10-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/on-curriculum-learning-for-commonsense","slug":"on-curriculum-learning-for-commonsense","title":"On Curriculum Learning for Commonsense Reasoning","date":"2022-07-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/when-chosen-wisely-more-data-is-what-you-need-1","slug":"when-chosen-wisely-more-data-is-what-you-need-1","title":"When Chosen Wisely, More Data Is What You Need: A Universal Sample-Efficient Strategy For Data Augmentation","date":"2022-03-17","arxiv_id":"2203.09391","repositories_listed":1,"syntology":null},{"url":"/paper/unicorn-on-rainbow-a-universal-commonsense","slug":"unicorn-on-rainbow-a-universal-commonsense","title":"UNICORN on RAINBOW: A Universal Commonsense Reasoning Model on a New Multitask Benchmark","date":"2021-03-24","arxiv_id":"2103.13009","repositories_listed":1,"syntology":{"n":5,"n_ran":3,"n_constructed":0,"n_ran_checked":3,"n_instrument":0,"n_unverified":2,"n_honours":0,"n_violates":0,"n_no_contract":3,"n_pointer_only":0,"phrase":"3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified","sample_list":"/paper/unicorn-on-rainbow-a-universal-commonsense#ran","syntology_url":"https://syntology.ai/paper/2103.13009","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2103.13009"}},"official":{"repos":["allenai/rainbow"],"state":"official (archive's flag): 3 ran","n_ran":3,"n_constructed":0,"n_ran_no_instrument_failure":3,"n_unverified":2,"ran_from_kinds":["official"]}}},{"url":null,"slug":"slimming-down-llms-without-losing-their-minds","title":"Slimming Down LLMs Without Losing Their Minds","date":"2025-06-12","arxiv_id":"2506.10885","repositories_listed":0,"syntology":null},{"url":null,"slug":"self-reasoning-language-models-unfold-hidden","title":"Self-Reasoning Language Models: Unfold Hidden Reasoning Chains with Few Reasoning Catalyst","date":"2025-05-20","arxiv_id":"2505.14116","repositories_listed":0,"syntology":null},{"url":null,"slug":"elastic-weight-consolidation-for-full","title":"Elastic Weight Consolidation for Full-Parameter Continual Pre-Training of Gemma2","date":"2025-05-09","arxiv_id":"2505.05946","repositories_listed":0,"syntology":null},{"url":null,"slug":"domain-adaptive-continued-pre-training-of","title":"Domain-Adaptive Continued Pre-Training of Small Language Models","date":"2025-04-13","arxiv_id":"2504.09687","repositories_listed":0,"syntology":null},{"url":null,"slug":"more-is-less-the-pitfalls-of-multi-model","title":"More is Less: The Pitfalls of Multi-Model Synthetic Preference Data in DPO Safety Alignment","date":"2025-04-03","arxiv_id":"2504.02193","repositories_listed":0,"syntology":null},{"url":null,"slug":"obliviate-efficient-unmemorization-for","title":"Obliviate: Efficient Unmemorization for Protecting Intellectual Property in Large Language Models","date":"2025-02-20","arxiv_id":"2502.15010","repositories_listed":0,"syntology":null},{"url":null,"slug":"hellaswag-pro-a-large-scale-bilingual","title":"HellaSwag-Pro: A Large-Scale Bilingual Benchmark for Evaluating the Robustness of LLMs in Commonsense Reasoning","date":"2025-02-17","arxiv_id":"2502.11393","repositories_listed":0,"syntology":null},{"url":null,"slug":"smoltulu-higher-learning-rate-to-batch-size","title":"SmolTulu: Higher Learning Rate to Batch Size Ratios Can Lead to Better Reasoning in SLMs","date":"2024-12-11","arxiv_id":"2412.08347","repositories_listed":0,"syntology":null},{"url":null,"slug":"towards-cross-lingual-llm-evaluation-for","title":"Towards Multilingual LLM Evaluation for European Languages","date":"2024-10-11","arxiv_id":"2410.08928","repositories_listed":0,"syntology":null},{"url":null,"slug":"progress-report-towards-european-llms","title":"Teuken-7B-Base & Teuken-7B-Instruct: Towards European LLMs","date":"2024-09-30","arxiv_id":"2410.03730","repositories_listed":0,"syntology":null},{"url":null,"slug":"grin-gradient-informed-moe","title":"GRIN: GRadient-INformed MoE","date":"2024-09-18","arxiv_id":"2409.12136","repositories_listed":0,"syntology":null},{"url":null,"slug":"promises-outlooks-and-challenges-of-diffusion","title":"Promises, Outlooks and Challenges of Diffusion Language Modeling","date":"2024-06-17","arxiv_id":"2406.11473","repositories_listed":0,"syntology":null},{"url":null,"slug":"who-s-harry-potter-approximate-unlearning-in","title":"Who's Harry Potter? Approximate Unlearning in LLMs","date":"2023-10-03","arxiv_id":"2310.02238","repositories_listed":0,"syntology":null},{"url":null,"slug":"contrastive-decoding-improves-reasoning-in","title":"Contrastive Decoding Improves Reasoning in Large Language Models","date":"2023-09-17","arxiv_id":"2309.09117","repositories_listed":0,"syntology":null},{"url":null,"slug":"when-chosen-wisely-more-data-is-what-you-need","title":"When Chosen Wisely, More Data Is What You Need: A Universal Sample-Efficient Strategy For Data Augmentation","date":"2021-11-16","arxiv_id":null,"repositories_listed":0,"syntology":null},{"url":null,"slug":"comparing-test-sets-with-item-response-theory","title":"Comparing Test Sets with Item Response Theory","date":"2021-06-01","arxiv_id":"2106.00840","repositories_listed":0,"syntology":null},{"url":"/paper/english-intermediate-task-training-improves","slug":"english-intermediate-task-training-improves","title":"English Intermediate-Task Training Improves Zero-Shot Cross-Lingual Transfer Too","date":"2020-05-26","arxiv_id":"2005.13013","repositories_listed":0,"syntology":null},{"url":null,"slug":"pre-training-is-almost-all-you-need-an","title":"Pre-training Is (Almost) All You Need: An Application to Commonsense Reasoning","date":"2020-04-29","arxiv_id":"2004.14074","repositories_listed":0,"syntology":null}],"syntology_records":8,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"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)"},"since_archive_judged":{"papers":0,"papers_placed":0,"judged_by":{"model":"a model","models_as_recorded":[],"from":"lb_rows_run_verified.jsonl: each record's `model` field (the extractor's call that wrote decisions[])"},"note":"a model's judgement from the paper's own tables that they report on this leaderboard, not a result: no value from a paper that was not placed is printed or carried here; papers counts each paper once over these leaderboards","backfill":"8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read)","backfill_complete":false,"leaderboards":[]},"papers_where_code_ran":8,"papers_where_code_ran_first_page":"/task/hellaswag/papers/ran/1","papers_where_code_ran_caption":"We ran code from the paper's repository; we did not run it on this task or check it against the task's benchmarks.","pages":[{"order":"ran","page":1,"of":1,"url":"/task/hellaswag/papers/ran/1","json":"/task/hellaswag/papers/ran/1.json","rows":8}],"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-10-05T00:36:11+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"},"row_route":"each syntology object on a benchmark's first row and on a paper carries sample_list (this site's paper page, #ran), syntology_url (syntology.ai's page for the paper, only when that route probed live) and mcp (the agent call that lists the paper's harvested samples; not a verdict on the paper)","developers":"https://syntology.ai/developers"},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"},"record_sha256":"0af96439d4213003c4c955f8fccf701f2041b3d04cf101097f187512ac1bec27","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}