{"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/how-predictable-are-large-language-model","title":"How Predictable Are Large Language Model Capabilities? A Case Study on BIG-bench","arxiv_id":"2305.14947","date":"2023-05-24","proceeding":null,"authors":["Qinyuan Ye","Harvey Yiyun Fu","Xiang Ren","Robin Jia"],"abstract":"We investigate the predictability of large language model (LLM) capabilities: given records of past experiments using different model families, numbers of parameters, tasks, and numbers of in-context examples, can we accurately predict LLM performance on new experiment configurations? Answering this question has practical implications for LLM users (e.g., deciding which models to try), developers (e.g., prioritizing evaluation on representative tasks), and the research community (e.g., identifying hard-to-predict capabilities that warrant further investigation). We study the performance prediction problem on experiment records from BIG-bench. On a random train-test split, an MLP-based predictor achieves an $R^2$ score greater than 95%, indicating the presence of learnable patterns within the experiment records. We then formulate the problem of searching for \"small-bench,\" an informative subset of BIG-bench tasks from which the performance on the full set can be maximally recovered. We find a subset as informative as BIG-bench Hard for evaluating new model families, while being $3\\times$ smaller. Additionally, we find competitive subsets by clustering task representations learned by our MLP-based predictor and selecting tasks close to cluster centroids, highlighting the importance of task diversity in constructing \"small-bench.\"","url_abs":"https://arxiv.org/abs/2305.14947v2","url_pdf":"https://arxiv.org/pdf/2305.14947v2.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":"how-predictable-are-large-language-model","repo_url":"https://github.com/ink-usc/predicting-big-bench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.14947","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.14947"}},"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/ink-usc/predicting-big-bench","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":15},"by_repo_kind":{"official":{"samples":15,"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":0,"samples":[{"code_sha256_prefix":"8a47bb5c2d961c22","entry":"crossover","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/search/genetic_algorithm.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/search/genetic_algorithm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8a47bb5c2d961c22"}},{"code_sha256_prefix":"89ca1fdd7d6460f3","entry":"eval_","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/mlp_models.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/mlp_models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"89ca1fdd7d6460f3"}},{"code_sha256_prefix":"f92dad82a57ddd5a","entry":"fill_random_to_length","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/search/genetic_algorithm.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/search/genetic_algorithm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f92dad82a57ddd5a"}},{"code_sha256_prefix":"1940d9b8ebfc16e7","entry":"generate_random_combinations","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/utils.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1940d9b8ebfc16e7"}},{"code_sha256_prefix":"1a2bfcf44ccd41c5","entry":"mutation","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/search/genetic_algorithm.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/search/genetic_algorithm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1a2bfcf44ccd41c5"}},{"code_sha256_prefix":"9be5b16b823daf48","entry":"reorder_surprise_preds","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/surprise_utils.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/surprise_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9be5b16b823daf48"}},{"code_sha256_prefix":"03fa5d9f7184220a","entry":"run_mlp_regression","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/mlp.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/mlp.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"03fa5d9f7184220a"}},{"code_sha256_prefix":"0d0a8acf4803a219","entry":"run_mlp_regression_for_search","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/mlp.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/mlp.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0d0a8acf4803a219"}},{"code_sha256_prefix":"d6ea72044eb758bc","entry":"run_random_forest","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/random_forest.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/random_forest.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d6ea72044eb758bc"}},{"code_sha256_prefix":"791ba96633405dff","entry":"run_xgb_regressor","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/xgb_regressor.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/xgb_regressor.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"791ba96633405dff"}},{"code_sha256_prefix":"314ea4df358695cc","entry":"train","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/mlp_models.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/mlp_models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"314ea4df358695cc"}},{"code_sha256_prefix":"adbfdafafb77c80f","entry":"train_for_search","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/mlp_models.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/mlp_models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"adbfdafafb77c80f"}},{"code_sha256_prefix":"72442426f6ca119f","entry":"tune_mlp_regression","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/mlp.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/mlp.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"72442426f6ca119f"}},{"code_sha256_prefix":"56dc5cc27ac2b05e","entry":"tune_random_forest","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/random_forest.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/random_forest.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"56dc5cc27ac2b05e"}},{"code_sha256_prefix":"11af8b257e4dafc7","entry":"tune_xgb_regressor","repo":"ink-usc/predicting-big-bench","repo_kind":"official","path":"code/models/xgb_regressor.py","file_url":"https://github.com/ink-usc/predicting-big-bench/blob/HEAD/code/models/xgb_regressor.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"11af8b257e4dafc7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}