{"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/arxiv-2511-00351","title":"Reject Only Critical Tokens: Pivot-Aware Speculative Decoding","arxiv_id":"2511.00351","date":"2025-11-01","proceeding":"NeurIPS","authors":["Amir Ziashahabi","Yavuz Faruk Bakman","Duygu Nur Yaldiz","Mostafa El-Khamy","Sai Praneeth Karimireddy","Salman Avestimehr"],"abstract":"Speculative Decoding (SD) ensures that the output matches the target model's distribution exactly. However, we argue that this distribution matching requirement is too stringent and results in unnecessarily low acceptance rates, limiting potential speedups. Instead, we advocate a reformulation of the decoding objective: the proposed decoding strategy should match the expected utility, i.e., the task-specific performance, of the target model. This perspective also aligns better with real-world use cases of LLMs, where utility (e.g., code correctness, factual accuracy) is often more important than sampling distribution. Based on this reformulation, we propose a novel decoding strategy: Pivot-Aware Speculative Decoding, which rejects only those tokens that would lead to a utility drop in the final output. We refer to these critical tokens as pivot tokens. We propose a method for labeling tokens as pivotal or non-pivotal and train a lightweight classifier to detect them. This method can be viewed as a relaxed version of standard SD, which offers much higher acceptance while preserving utility. We evaluate our method across various datasets, demonstrating that we can achieve up to $2.5\\times$ speedup with comparable utility. Source code is available at https://github.com/amir-zsh/PAD.","url_abs":"https://arxiv.org/abs/2511.00351","url_pdf":"https://arxiv.org/pdf/2511.00351","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2511.00351","atlas_url":"https://app.syntology.ai/?focus=2511.00351","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2511.00351"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/amir-zsh/PAD","reach":{"status":"ok"}}],"summary":{"ran_honours":1,"ran_draft_wrong":1,"unverified":14},"by_repo_kind":{"found_in_text":{"samples":16,"ran":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":16,"samples":[{"code_sha256_prefix":"aba86ffbed27ba65","entry":"find_multiple","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/model.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/model.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"aba86ffbed27ba65"}},{"code_sha256_prefix":"0b5f90e8856ad7af","entry":"multinomial_sample_one_no_sync","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/gpt_fast_utils.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/gpt_fast_utils.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0b5f90e8856ad7af"}},{"code_sha256_prefix":"e32aa65ed977997d","entry":"apply_rope_scaling","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/model.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e32aa65ed977997d"}},{"code_sha256_prefix":"ebe7962c6989540f","entry":"apply_top_k_only","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/gpt_fast_utils.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/gpt_fast_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ebe7962c6989540f"}},{"code_sha256_prefix":"6d9be873169e847a","entry":"apply_top_k_top_p","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/gpt_fast_utils.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/gpt_fast_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6d9be873169e847a"}},{"code_sha256_prefix":"3ef18ccd1206347c","entry":"convert_act_dict_to_tensors","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"train_classifier.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/train_classifier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3ef18ccd1206347c"}},{"code_sha256_prefix":"55ec8fea60e1d7e7","entry":"dynamically_quantize_per_channel","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/quantize.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/quantize.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"55ec8fea60e1d7e7"}},{"code_sha256_prefix":"ce2b1e2625c49910","entry":"eval","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"train_classifier.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/train_classifier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ce2b1e2625c49910"}},{"code_sha256_prefix":"dbabfb89a4b530ac","entry":"get_group_qparams","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/quantize.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/quantize.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dbabfb89a4b530ac"}},{"code_sha256_prefix":"f4da6be2e1f446cf","entry":"get_logits_and_embeddings","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"train_classifier.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/train_classifier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f4da6be2e1f446cf"}},{"code_sha256_prefix":"36cf05d05fbba908","entry":"get_tokenizer","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/tokenizer.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/tokenizer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"36cf05d05fbba908"}},{"code_sha256_prefix":"914469751a0f1948","entry":"load_mlp_model","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"decode.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/decode.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"914469751a0f1948"}},{"code_sha256_prefix":"146cdcef0c195fd7","entry":"pack_scales_and_zeros","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"src/gpt_fast/quantize.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/src/gpt_fast/quantize.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"146cdcef0c195fd7"}},{"code_sha256_prefix":"c1e4d88b26b6d460","entry":"select_kv_cache_portion","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"dataset_generation.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/dataset_generation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c1e4d88b26b6d460"}},{"code_sha256_prefix":"b97cd418550d173e","entry":"str2bool","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"decode.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/decode.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b97cd418550d173e"}},{"code_sha256_prefix":"86c8ddad5eea6789","entry":"strike","repo":"amir-zsh/PAD","repo_kind":"found_in_text","path":"decode.py","file_url":"https://github.com/amir-zsh/PAD/blob/HEAD/decode.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"86c8ddad5eea6789"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_api"},"syntology_extracted_results":null}