{"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/sloth-scaling-laws-for-llm-skills-to-predict","title":"Sloth: scaling laws for LLM skills to predict multi-benchmark performance across families","arxiv_id":"2412.06540","date":"2024-12-09","proceeding":null,"authors":["Felipe Maia Polo","Seamus Somerstep","Leshem Choshen","Yuekai Sun","Mikhail Yurochkin"],"abstract":"Scaling laws for large language models (LLMs) predict model performance based on parameters like size and training data. However, differences in training configurations and data processing across model families lead to significant variations in benchmark performance, making it difficult for a single scaling law to generalize across all LLMs. On the other hand, training family-specific scaling laws requires training models of varying sizes for every family. In this work, we propose Skills Scaling Laws (SSLaws, pronounced as Sloth), a novel scaling law that leverages publicly available benchmark data and assumes LLM performance is driven by low-dimensional latent skills, such as reasoning and instruction following. These latent skills are influenced by computational resources like model size and training tokens but with varying efficiencies across model families. Sloth exploits correlations across benchmarks to provide more accurate and interpretable predictions while alleviating the need to train multiple LLMs per family. We present both theoretical results on parameter identification and empirical evaluations on 12 prominent benchmarks, from Open LLM Leaderboard v1/v2, demonstrating that Sloth predicts LLM performance efficiently and offers insights into scaling behaviors for downstream tasks such as coding and emotional intelligence applications.","url_abs":"https://arxiv.org/abs/2412.06540v3","url_pdf":"https://arxiv.org/pdf/2412.06540v3.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":"sloth-scaling-laws-for-llm-skills-to-predict","repo_url":"https://github.com/felipemaiapolo/sloth","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"emotional-intelligence","task_name":"Emotional Intelligence"},{"task_slug":"instruction-following","task_name":"Instruction Following"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2412.06540","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2412.06540"}},"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":"deterministic:regex_extraction","url":"https://github.com/felipemaiapolo/sloth","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":9},"by_repo_kind":{"official":{"samples":9,"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":"869a887d132f2470","entry":"filter","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/utils.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"869a887d132f2470"}},{"code_sha256_prefix":"83a97028835dd404","entry":"forward1","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/sloth.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/sloth.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"83a97028835dd404"}},{"code_sha256_prefix":"f975285e09aeea1e","entry":"forward2","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/sloth.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/sloth.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f975285e09aeea1e"}},{"code_sha256_prefix":"94abc955695450b9","entry":"get_true_indices","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/utils.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"94abc955695450b9"}},{"code_sha256_prefix":"c96121833ed7e023","entry":"logit","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/sloth.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/sloth.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c96121833ed7e023"}},{"code_sha256_prefix":"004484c84acb696d","entry":"prep_data","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/experiment.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/experiment.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"004484c84acb696d"}},{"code_sha256_prefix":"a9375f45a421c52a","entry":"prep_data2","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/experiment.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/experiment.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a9375f45a421c52a"}},{"code_sha256_prefix":"38473214bda25f09","entry":"run_exp","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/experiment.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/experiment.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"38473214bda25f09"}},{"code_sha256_prefix":"5a36846dab98ce35","entry":"sigmoid_np","repo":"felipemaiapolo/sloth","repo_kind":"official","path":"sloth/utils.py","file_url":"https://github.com/felipemaiapolo/sloth/blob/HEAD/sloth/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5a36846dab98ce35"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}