Papers › Breaking the Ceiling of the LLM Community by Treating Token Generation as a...

Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling

18 Jun 2024arXiv:2406.12585archive 2025-07-28

Yao-Ching Yu, Chun-Chih Kuo, Ziqi Ye, Yu-Cheng Chang, Yueh-Se Li

Ensembling multiple models has always been an effective approach to push the limits of existing performance and is widely used in classification tasks by simply averaging the classification probability vectors from multiple classifiers to achieve better accuracy. However, in the thriving open-source Large Language Model (LLM) community, ensembling methods are rare and typically limited to ensembling the full-text outputs of LLMs, such as selecting the best output using a ranker, which leads to underutilization of token-level probability information. In this paper, we treat the Generation of each token by LLMs as a Classification (GaC) for ensembling. This approach fully exploits the probability information at each generation step and better prevents LLMs from producing early incorrect tokens that lead to snowballing errors. In experiments, we ensemble state-of-the-art LLMs on several benchmarks, including exams, mathematics and reasoning, and observe that our method breaks the existing community performance ceiling. Furthermore, we observed that most of the tokens in the answer are simple and do not affect the correctness of the final answer. Therefore, we also experimented with ensembling only key tokens, and the results showed better performance with lower latency across benchmarks.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2406.12585")

Code

Syntology Ran 4 of 7 code samples harvested from 1 repository linked to this paper; 3 have no recorded run. Of those that ran: 4 ran with no contract checked.

By repository: official repository: 7 samples from 1 repository, 4 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

yaoching0/gac officialmentioned in papermentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

7 samples harvested; 4 ran; 0 honoured the contract we drafted; 3 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

4ran
3unverified

Licence: 0 of the 7 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from yaoching0/gac. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

check_priorities yaoching0/gac/utils/gac_gen_call.py official repository ran MIT (permissive) · 250e3d0aab21f5dc · report
synchronize_unfinished_sequences yaoching0/gac/utils/gac_gen_utils.py official repository ran MIT (permissive) · 3b4f685120e04a9c · report
update_input_ids_and_model_kwargs yaoching0/gac/utils/gac_gen_utils.py official repository ran MIT (permissive) · a1d6799b4d0d673e · report
validate_and_update_quantization yaoching0/gac/utils/gac_gen_call.py official repository ran MIT (permissive) · 4fda25a3a31731c5 · report
get_remote_model_generator_class yaoching0/gac/utils/ray_actor.py official repository unverified MIT (permissive) · 5640e58ac0f3f8c3 · report
setup_custom_logger yaoching0/gac/utils/logger.py official repository unverified MIT (permissive) · e008364719672578 · report
setup_model_actors_and_data yaoching0/gac/utils/gac_gen_call.py official repository unverified MIT (permissive) · 03e59517a81a3729 · report

Tasks

Arithmetic ReasoningLanguage ModelingLanguage ModellingLarge Language ModelMulti-task Language UnderstandingQuestion Answering

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Arithmetic Reasoning GSM8K GaC(Qwen2-72B-Instruct + Llama-3-70B-Instruct) Accuracy 90.91 #19 of 164 Archive leaderboard report
Question Answering TriviaQA GaC(Qwen2-72B-Instruct + Llama-3-70B-Instruct) EM 79.29 #14 of 56 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections