{"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/information-entropy-invariance-enhancing","title":"Information Entropy Invariance: Enhancing Length Extrapolation in Attention Mechanisms","arxiv_id":"2501.08570","date":"2025-01-15","proceeding":null,"authors":["Kewei Li","Yanwen Kong","Yiping Xu","Jianlin Su","Lan Huang","Ruochi Zhang","Fengfeng Zhou"],"abstract":"Since the emergence of research on improving the length extrapolation capabilities of large language models in 2021, some studies have made modifications to the scaling factor in the scaled dot-product attention mechanism as part of their proposed methods without rigorous theoretical justifications. To fill this gap, we propose two new scaled temperatures based on information entropy invariance to enhance length extrapolation. First, a training-free method InfoScale is designed for dotproduct attention, and preserves focus on original tokens during length extrapolation by ensuring consistent entropy. Second, we theoretically analyze the impact of scaling (CosScale) on cosine attention. Experimental data demonstrates that combining InfoScale and CosScale achieves state-ofthe-art performance on the GAU-{\\alpha} model with a context window extended to 64 times the training length, and outperforms seven existing methods. Our analysis reveals that significantly increasing CosScale approximates the Windowed Attention, and highlights the significance of attention score dilution as a key challenge in long-range context handling. The code and data are available at https://github.com/HT-NEKO/ Information-Entropy-Invariance.","url_abs":"https://arxiv.org/abs/2501.08570v3","url_pdf":"https://arxiv.org/pdf/2501.08570v3.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":"information-entropy-invariance-enhancing","repo_url":"https://github.com/ht-neko/information-entropy-invariance","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"focus","method_name":"Focus"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2501.08570","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2501.08570"}},"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/ht-neko/information-entropy-invariance","reach":null}],"summary":{"ran_draft_wrong":3},"by_repo_kind":{"official":{"samples":3,"ran":3,"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":3,"samples":[{"code_sha256_prefix":"12c35a5891d87be1","entry":"generate_kmer_str","repo":"ht-neko/information-entropy-invariance","repo_kind":"official","path":"tasks/DNA/DNABERT_2/train_kyw.py","file_url":"https://github.com/ht-neko/information-entropy-invariance/blob/HEAD/tasks/DNA/DNABERT_2/train_kyw.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"12c35a5891d87be1"}},{"code_sha256_prefix":"e6b562614d5aa421","entry":"get_alter_of_dna_sequence","repo":"ht-neko/information-entropy-invariance","repo_kind":"official","path":"tasks/DNA/DNABERT_2/train_kyw.py","file_url":"https://github.com/ht-neko/information-entropy-invariance/blob/HEAD/tasks/DNA/DNABERT_2/train_kyw.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e6b562614d5aa421"}},{"code_sha256_prefix":"4adebeb37f990a23","entry":"load_or_generate_kmer","repo":"ht-neko/information-entropy-invariance","repo_kind":"official","path":"tasks/DNA/DNABERT_2/train_kyw.py","file_url":"https://github.com/ht-neko/information-entropy-invariance/blob/HEAD/tasks/DNA/DNABERT_2/train_kyw.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4adebeb37f990a23"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}