@inproceedings{vasu-etal-2026-harpa,
title = "{HARPA}: A Testability-Driven, Literature-Grounded Framework for Research Ideation",
author = "Vasu, Rosni and
Jansen, Peter and
Siangliulue, Pao and
Sarasua, Cristina and
Bernstein, Abraham and
Clark, Peter and
Dalvi Mishra, Bhavana",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.215/",
pages = "4821--4880",
ISBN = "979-8-89176-439-2",
abstract = "Recent advances in large language models (LLMs) have accelerated automated scientific discovery (ASD), but existing ideation tools still struggle to generate hypotheses that are both testable, literature-grounded, and adaptive to prior experimental outcomes. We developed HARPA to address these challenges by following a human-inspired ideation workflow that mines literature trends, explores hypothesis design spaces, and converges on precise, testable hypotheses by pinpointing research gaps and justifying design choices. Human evaluations show that HARPA-generated hypothesis-driven research proposals performs comparably to a strong baseline system, AI-researcher, in novelty and overall quality, while showing a positive trend in feasibility (+0.78, $p<0.05$, bootstrap) and a significant gain in groundedness (+0.85, $p<0.01$, bootstrap) on a 10-point Likert scale. When evaluated with the ASD agent CodeScientist, HARPA produced more successful executions (20 vs. 11 out of 40) and fewer failures (16 vs. 21 out of 40) indicating that expert feasibility judgments track execution success. HARPA also learns a reward model that scores hypotheses based on prior experimental outcomes, achieving approx. a 28{\%} absolute gain over HARPA{'}s untrained baseline scorer."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vasu-etal-2026-harpa">
<titleInfo>
<title>HARPA: A Testability-Driven, Literature-Grounded Framework for Research Ideation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rosni</namePart>
<namePart type="family">Vasu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peter</namePart>
<namePart type="family">Jansen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pao</namePart>
<namePart type="family">Siangliulue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cristina</namePart>
<namePart type="family">Sarasua</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abraham</namePart>
<namePart type="family">Bernstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peter</namePart>
<namePart type="family">Clark</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bhavana</namePart>
<namePart type="family">Dalvi Mishra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Recent advances in large language models (LLMs) have accelerated automated scientific discovery (ASD), but existing ideation tools still struggle to generate hypotheses that are both testable, literature-grounded, and adaptive to prior experimental outcomes. We developed HARPA to address these challenges by following a human-inspired ideation workflow that mines literature trends, explores hypothesis design spaces, and converges on precise, testable hypotheses by pinpointing research gaps and justifying design choices. Human evaluations show that HARPA-generated hypothesis-driven research proposals performs comparably to a strong baseline system, AI-researcher, in novelty and overall quality, while showing a positive trend in feasibility (+0.78, p<0.05, bootstrap) and a significant gain in groundedness (+0.85, p<0.01, bootstrap) on a 10-point Likert scale. When evaluated with the ASD agent CodeScientist, HARPA produced more successful executions (20 vs. 11 out of 40) and fewer failures (16 vs. 21 out of 40) indicating that expert feasibility judgments track execution success. HARPA also learns a reward model that scores hypotheses based on prior experimental outcomes, achieving approx. a 28% absolute gain over HARPA’s untrained baseline scorer.</abstract>
<identifier type="citekey">vasu-etal-2026-harpa</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.215/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>4821</start>
<end>4880</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HARPA: A Testability-Driven, Literature-Grounded Framework for Research Ideation
%A Vasu, Rosni
%A Jansen, Peter
%A Siangliulue, Pao
%A Sarasua, Cristina
%A Bernstein, Abraham
%A Clark, Peter
%A Dalvi Mishra, Bhavana
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F vasu-etal-2026-harpa
%X Recent advances in large language models (LLMs) have accelerated automated scientific discovery (ASD), but existing ideation tools still struggle to generate hypotheses that are both testable, literature-grounded, and adaptive to prior experimental outcomes. We developed HARPA to address these challenges by following a human-inspired ideation workflow that mines literature trends, explores hypothesis design spaces, and converges on precise, testable hypotheses by pinpointing research gaps and justifying design choices. Human evaluations show that HARPA-generated hypothesis-driven research proposals performs comparably to a strong baseline system, AI-researcher, in novelty and overall quality, while showing a positive trend in feasibility (+0.78, p<0.05, bootstrap) and a significant gain in groundedness (+0.85, p<0.01, bootstrap) on a 10-point Likert scale. When evaluated with the ASD agent CodeScientist, HARPA produced more successful executions (20 vs. 11 out of 40) and fewer failures (16 vs. 21 out of 40) indicating that expert feasibility judgments track execution success. HARPA also learns a reward model that scores hypotheses based on prior experimental outcomes, achieving approx. a 28% absolute gain over HARPA’s untrained baseline scorer.
%U https://aclanthology.org/2026.aacl-long.215/
%P 4821-4880
Markdown (Informal)
[HARPA: A Testability-Driven, Literature-Grounded Framework for Research Ideation](https://aclanthology.org/2026.aacl-long.215/) (Vasu et al., AACL-IJCNLP 2026)
ACL
- Rosni Vasu, Peter Jansen, Pao Siangliulue, Cristina Sarasua, Abraham Bernstein, Peter Clark, and Bhavana Dalvi Mishra. 2026. HARPA: A Testability-Driven, Literature-Grounded Framework for Research Ideation. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4821–4880, Hengqin, China. Association for Computational Linguistics.