@inproceedings{thakur-etal-2026-warp,
title = "{WARP}: {W}asserstein-Aligned {RAG} for Population Opinions",
author = "Thakur, Aman Singh and
Agrawal, Aditya and
Nakkiran, Alwarappan and
Karlsson, Alex",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.158/",
pages = "2684--2704",
ISBN = "979-8-89176-460-6",
abstract = "RAG systems are increasingly used to summarize what large collections of documents say. A user asks ``What do people think about X?'' and receives an answer that reads as consensus. But standard top-k retrieval ranks documents by query similarity, not by how faithfully they represent the population, so minority views quietly disappear. Existing fixes fall short. Diversity re-rankers like MMR and DPP spread retrieved documents apart, but with no target distribution to aim for. Calibration methods based on KL or JS divergence do target one, yet treat opinion bins as unordered: confusing strong positive with strong negative costs no more than an adjacent-bin miss.\par We introduce WARP, a family of post-retrieval algorithms that calibrate retrieved evidence to the population{'}s opinion distribution. WARP first recovers underrepresented opinions that cosine ranking may bury, then uses Wasserstein-1 distance to select documents whose sentiment-intensity distribution matches the population target, capturing the ordinal structure ignored by KL and JS divergence. We develop three variants for dense, sparse, and variable candidate pools, trading off calibration quality and speed. Across three review domains spanning 35K documents, 156 queries, and 26 entities, WARP{'}s domain-matched variants reduce distributional error by at least 43{\%} with sub-second latency. These gains carry through to generation: a five-judge LLM panel prefers WARP-generated answers in up to 86{\%} of decided comparisons at k=5."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="thakur-etal-2026-warp">
<titleInfo>
<title>WARP: Wasserstein-Aligned RAG for Population Opinions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aman</namePart>
<namePart type="given">Singh</namePart>
<namePart type="family">Thakur</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aditya</namePart>
<namePart type="family">Agrawal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alwarappan</namePart>
<namePart type="family">Nakkiran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alex</namePart>
<namePart type="family">Karlsson</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</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">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>RAG systems are increasingly used to summarize what large collections of documents say. A user asks “What do people think about X?” and receives an answer that reads as consensus. But standard top-k retrieval ranks documents by query similarity, not by how faithfully they represent the population, so minority views quietly disappear. Existing fixes fall short. Diversity re-rankers like MMR and DPP spread retrieved documents apart, but with no target distribution to aim for. Calibration methods based on KL or JS divergence do target one, yet treat opinion bins as unordered: confusing strong positive with strong negative costs no more than an adjacent-bin miss.\par We introduce WARP, a family of post-retrieval algorithms that calibrate retrieved evidence to the population’s opinion distribution. WARP first recovers underrepresented opinions that cosine ranking may bury, then uses Wasserstein-1 distance to select documents whose sentiment-intensity distribution matches the population target, capturing the ordinal structure ignored by KL and JS divergence. We develop three variants for dense, sparse, and variable candidate pools, trading off calibration quality and speed. Across three review domains spanning 35K documents, 156 queries, and 26 entities, WARP’s domain-matched variants reduce distributional error by at least 43% with sub-second latency. These gains carry through to generation: a five-judge LLM panel prefers WARP-generated answers in up to 86% of decided comparisons at k=5.</abstract>
<identifier type="citekey">thakur-etal-2026-warp</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.158/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>2684</start>
<end>2704</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T WARP: Wasserstein-Aligned RAG for Population Opinions
%A Thakur, Aman Singh
%A Agrawal, Aditya
%A Nakkiran, Alwarappan
%A Karlsson, Alex
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F thakur-etal-2026-warp
%X RAG systems are increasingly used to summarize what large collections of documents say. A user asks “What do people think about X?” and receives an answer that reads as consensus. But standard top-k retrieval ranks documents by query similarity, not by how faithfully they represent the population, so minority views quietly disappear. Existing fixes fall short. Diversity re-rankers like MMR and DPP spread retrieved documents apart, but with no target distribution to aim for. Calibration methods based on KL or JS divergence do target one, yet treat opinion bins as unordered: confusing strong positive with strong negative costs no more than an adjacent-bin miss.\par We introduce WARP, a family of post-retrieval algorithms that calibrate retrieved evidence to the population’s opinion distribution. WARP first recovers underrepresented opinions that cosine ranking may bury, then uses Wasserstein-1 distance to select documents whose sentiment-intensity distribution matches the population target, capturing the ordinal structure ignored by KL and JS divergence. We develop three variants for dense, sparse, and variable candidate pools, trading off calibration quality and speed. Across three review domains spanning 35K documents, 156 queries, and 26 entities, WARP’s domain-matched variants reduce distributional error by at least 43% with sub-second latency. These gains carry through to generation: a five-judge LLM panel prefers WARP-generated answers in up to 86% of decided comparisons at k=5.
%U https://aclanthology.org/2026.emnlp-industry.158/
%P 2684-2704
Markdown (Informal)
[WARP: Wasserstein-Aligned RAG for Population Opinions](https://aclanthology.org/2026.emnlp-industry.158/) (Thakur et al., EMNLP 2026)
ACL
- Aman Singh Thakur, Aditya Agrawal, Alwarappan Nakkiran, and Alex Karlsson. 2026. WARP: Wasserstein-Aligned RAG for Population Opinions. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 2684–2704, Budapest, Hungary. Association for Computational Linguistics.