@inproceedings{lorandi-etal-2026-automatic,
title = "Automatic Paper Analysis and Categorisation for Systematic Reviews with Combined Reasoning-Augmented {SFT} and {DAPO} {RL}",
author = "Lorandi, Michela and
Belz, Anya and
Mille, Simon and
Thomson, Craig",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.2022/",
pages = "40676--40699",
ISBN = "979-8-89176-395-1",
abstract = "Systematic reviews are a cornerstone of modern science, synthesising evidence from published research to provide the highest level of research evidence in a field. The process includes categorising studies on a number of different dimensions which is laborious and time consuming. Automatic approaches are beginning to be explored but the complexity of the task means we are currently far from a satisfactory solution. In this paper, we test different annotation scheme-agnostic methods for automatic NLP paper categorisation for systematic reviews, and test them on two tasks: (i) annotating NLP papers for categories of reported controlled-text generation methods, and (ii) annotating NLP papers for categories of reported human evaluations. We find that reasoning-enhanced fine-tuning combined with DAPO reinforcement learning rewarding both correctness and output format substantially improves the performance of LLMs (by up to +53.8 points), even when they have been pre-trained to perform reasoning, and cuts time required for annotation by around 80{\%} in a human-in-the-loop setting."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lorandi-etal-2026-automatic">
<titleInfo>
<title>Automatic Paper Analysis and Categorisation for Systematic Reviews with Combined Reasoning-Augmented SFT and DAPO RL</title>
</titleInfo>
<name type="personal">
<namePart type="given">Michela</namePart>
<namePart type="family">Lorandi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anya</namePart>
<namePart type="family">Belz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Mille</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Craig</namePart>
<namePart type="family">Thomson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Systematic reviews are a cornerstone of modern science, synthesising evidence from published research to provide the highest level of research evidence in a field. The process includes categorising studies on a number of different dimensions which is laborious and time consuming. Automatic approaches are beginning to be explored but the complexity of the task means we are currently far from a satisfactory solution. In this paper, we test different annotation scheme-agnostic methods for automatic NLP paper categorisation for systematic reviews, and test them on two tasks: (i) annotating NLP papers for categories of reported controlled-text generation methods, and (ii) annotating NLP papers for categories of reported human evaluations. We find that reasoning-enhanced fine-tuning combined with DAPO reinforcement learning rewarding both correctness and output format substantially improves the performance of LLMs (by up to +53.8 points), even when they have been pre-trained to perform reasoning, and cuts time required for annotation by around 80% in a human-in-the-loop setting.</abstract>
<identifier type="citekey">lorandi-etal-2026-automatic</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.2022/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>40676</start>
<end>40699</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Automatic Paper Analysis and Categorisation for Systematic Reviews with Combined Reasoning-Augmented SFT and DAPO RL
%A Lorandi, Michela
%A Belz, Anya
%A Mille, Simon
%A Thomson, Craig
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F lorandi-etal-2026-automatic
%X Systematic reviews are a cornerstone of modern science, synthesising evidence from published research to provide the highest level of research evidence in a field. The process includes categorising studies on a number of different dimensions which is laborious and time consuming. Automatic approaches are beginning to be explored but the complexity of the task means we are currently far from a satisfactory solution. In this paper, we test different annotation scheme-agnostic methods for automatic NLP paper categorisation for systematic reviews, and test them on two tasks: (i) annotating NLP papers for categories of reported controlled-text generation methods, and (ii) annotating NLP papers for categories of reported human evaluations. We find that reasoning-enhanced fine-tuning combined with DAPO reinforcement learning rewarding both correctness and output format substantially improves the performance of LLMs (by up to +53.8 points), even when they have been pre-trained to perform reasoning, and cuts time required for annotation by around 80% in a human-in-the-loop setting.
%U https://aclanthology.org/2026.findings-acl.2022/
%P 40676-40699
Markdown (Informal)
[Automatic Paper Analysis and Categorisation for Systematic Reviews with Combined Reasoning-Augmented SFT and DAPO RL](https://aclanthology.org/2026.findings-acl.2022/) (Lorandi et al., Findings 2026)
ACL