@inproceedings{wang-etal-2026-rewardpartial,
title = "{R}eward{P}artial: A Completion Potential Evaluation Framework for Truncated Best-of-N Paradigm",
author = "Wang, Xin and
Wang, Yidong and
Wang, Qiufeng and
Jia, Bo and
Niu, Yilin and
Tang, Jie and
Wang, Cunxiang",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.488/",
pages = "10220--10234",
ISBN = "979-8-89176-462-0",
abstract = "Traditional decoding-time alignment methods, such as Best-of-N sampling, enhance output quality by selecting high-reward responses from multiple candidates. However, these approaches incur substantial computational overhead, which limits their practicality. Recent advances aim to mitigate this inefficiency by scoring and filtering partial generations early in the decoding process. However, these methods face a critical challenge: conventional reward models (RMs), optimized to assess complete responses, exhibit poor performance when evaluating truncated text. To address this, we introduce RewardPartial, a novel evaluation paradigm that shifts the focus from superficial partial-text quality to estimating a response{'}s completion potential. We provide theoretical analysis for this paradigm through the perspective of Markov Decision Process and construct TruncEval, a specialized dataset for training RewardPartial models to accurately predict this potential. Our evaluation on the TruncEval test set demonstrates that RewardPartial models deliver significant performance gains over conventional reward models. When deployed in early-filtering pipelines, our method achieves superior alignment performance compared to Best-of-N sampling, offering a more efficient solution for high-quality text generation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-rewardpartial">
<titleInfo>
<title>RewardPartial: A Completion Potential Evaluation Framework for Truncated Best-of-N Paradigm</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yidong</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qiufeng</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</namePart>
<namePart type="family">Jia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yilin</namePart>
<namePart type="family">Niu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jie</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cunxiang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Traditional decoding-time alignment methods, such as Best-of-N sampling, enhance output quality by selecting high-reward responses from multiple candidates. However, these approaches incur substantial computational overhead, which limits their practicality. Recent advances aim to mitigate this inefficiency by scoring and filtering partial generations early in the decoding process. However, these methods face a critical challenge: conventional reward models (RMs), optimized to assess complete responses, exhibit poor performance when evaluating truncated text. To address this, we introduce RewardPartial, a novel evaluation paradigm that shifts the focus from superficial partial-text quality to estimating a response’s completion potential. We provide theoretical analysis for this paradigm through the perspective of Markov Decision Process and construct TruncEval, a specialized dataset for training RewardPartial models to accurately predict this potential. Our evaluation on the TruncEval test set demonstrates that RewardPartial models deliver significant performance gains over conventional reward models. When deployed in early-filtering pipelines, our method achieves superior alignment performance compared to Best-of-N sampling, offering a more efficient solution for high-quality text generation.</abstract>
<identifier type="citekey">wang-etal-2026-rewardpartial</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.488/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>10220</start>
<end>10234</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T RewardPartial: A Completion Potential Evaluation Framework for Truncated Best-of-N Paradigm
%A Wang, Xin
%A Wang, Yidong
%A Wang, Qiufeng
%A Jia, Bo
%A Niu, Yilin
%A Tang, Jie
%A Wang, Cunxiang
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F wang-etal-2026-rewardpartial
%X Traditional decoding-time alignment methods, such as Best-of-N sampling, enhance output quality by selecting high-reward responses from multiple candidates. However, these approaches incur substantial computational overhead, which limits their practicality. Recent advances aim to mitigate this inefficiency by scoring and filtering partial generations early in the decoding process. However, these methods face a critical challenge: conventional reward models (RMs), optimized to assess complete responses, exhibit poor performance when evaluating truncated text. To address this, we introduce RewardPartial, a novel evaluation paradigm that shifts the focus from superficial partial-text quality to estimating a response’s completion potential. We provide theoretical analysis for this paradigm through the perspective of Markov Decision Process and construct TruncEval, a specialized dataset for training RewardPartial models to accurately predict this potential. Our evaluation on the TruncEval test set demonstrates that RewardPartial models deliver significant performance gains over conventional reward models. When deployed in early-filtering pipelines, our method achieves superior alignment performance compared to Best-of-N sampling, offering a more efficient solution for high-quality text generation.
%U https://aclanthology.org/2026.findings-emnlp.488/
%P 10220-10234
Markdown (Informal)
[RewardPartial: A Completion Potential Evaluation Framework for Truncated Best-of-N Paradigm](https://aclanthology.org/2026.findings-emnlp.488/) (Wang et al., Findings 2026)
ACL