arxiv: 2604.18122 · v1 · submitted 2026-04-20 · 💻 cs.CL

Recognition: unknown

Decisive: Guiding User Decisions with Optimal Preference Elicitation from Unstructured Documents

Aishani Pandey, Akriti Jain, Anish Mulay, Aparna Garimella, Divyansh Verma, Pritika Ramu

Authors on Pith no claims yet

Pith reviewed 2026-05-10 05:08 UTC · model grok-4.3

classification 💻 cs.CL

keywords decision makingpreference elicitationBayesian inferenceunstructured documentsinteractive systemspersonalizationlarge language models

0 comments

The pith

Decisive improves decision accuracy up to 20% by extracting option scores from documents and adaptively learning user preferences through targeted pairwise questions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents Decisive as an interactive framework that first builds an objective scoring matrix for decision options by processing unstructured source documents. It then maintains a low-dimensional latent vector representing the user's preferences and updates that vector by asking pairwise tradeoff questions chosen to maximize information gain about the final recommendation. This combination keeps recommendations grounded in the original documents while reducing the number of questions needed for personalization. Experiments across domains show the method outperforms both general large language models and prior decision-support systems.

Core claim

The central claim is that grounding decisions in a document-derived option-scoring matrix and updating a latent preference vector via information-gain-maximizing pairwise elicitations produces more accurate, transparent, and personalized recommendations than existing LLM-based or decision-support approaches.

What carries the argument

The adaptive elicitation module that selects pairwise questions to maximize expected information gain over the final decision, paired with Bayesian updates to the user's latent preference vector.

Load-bearing premise

The method assumes that an accurate and complete option-scoring matrix can be extracted from unstructured documents and that user preferences are adequately captured by a low-dimensional latent vector.

What would settle it

An experiment in which the automatically extracted scoring matrix contains systematic errors or in which real user preferences require more than the assumed number of dimensions, resulting in no accuracy gain over baselines.

Figures

Figures reproduced from arXiv: 2604.18122 by Aishani Pandey, Akriti Jain, Anish Mulay, Aparna Garimella, Divyansh Verma, Pritika Ramu.

**Figure 2.** Figure 2: Overview of the data generation process. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Overview of the DECISIVE pipeline. The system first extracts K preference factors from documents and constructs the Option-Scoring Matrix S, which quantifies how each option performs on each factor. A Dirichlet prior models uncertainty over the user’s preference weights, with particle samples representing plausible preference vectors. Decision-aware questions are adaptively selected to maximize information… view at source ↗

read the original abstract

Decision-making is a cognitively intensive task that requires synthesizing relevant information from multiple unstructured sources, weighing competing factors, and incorporating subjective user preferences. Existing methods, including large language models and traditional decision-support systems, fall short: they often overwhelm users with information or fail to capture nuanced preferences accurately. We present Decisive, an interactive decision-making framework that combines document-grounded reasoning with Bayesian preference inference. Our approach grounds decisions in an objective option-scoring matrix extracted from source documents, while actively learning a user's latent preference vector through targeted elicitation. Users answer pairwise tradeoff questions adaptively selected to maximize information gain over the final decision. This process converges efficiently, minimizing user effort while ensuring recommendations remain transparent and personalized. Through extensive experiments, we demonstrate that our approach significantly outperforms both general-purpose LLMs and existing decision-making frameworks achieving up to 20% improvement in decision accuracy over strong baselines across domains.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Decisive puts together document extraction and information-gain elicitation in a workable way, but the 20% accuracy gains rest on extraction quality that is not shown to be reliable.

read the letter

The paper introduces Decisive, which extracts an option-scoring matrix from source documents and then runs Bayesian preference inference with adaptively chosen pairwise questions. The specific loop—ground the scores in the documents first, then pick questions that maximize information gain on the final recommendation—does not appear in the prior work the abstract cites, so that combination counts as new. The write-up explains the flow clearly enough that a reader can see how the matrix feeds the latent vector and why the process stops early. That part is useful for anyone building interactive tools that must stay tied to real text rather than free-form LLM output. The transparency claim also follows directly from keeping the matrix visible to the user. The main weakness is the experimental support. The abstract states gains of up to 20% over strong baselines, yet supplies no description of how the extraction step was checked for completeness or accuracy, no inter-annotator numbers, and no ablation that adds noise to the matrix. If the extracted scores systematically drop important trade-offs or mis-weight conflicting passages, the later elicitation step is optimizing over the wrong space and the reported lift cannot be credited to the framework. The stress-test note flags exactly this gap, and nothing in the supplied text closes it. The work is aimed at people who design decision-support systems that ingest reports, reviews, or policy documents. A reader looking for a concrete recipe that mixes retrieval-style grounding with active preference learning will find the high-level design worth examining. I would send it to peer review. The core architecture is coherent and the practical motivation is clear; the experiments simply need the missing validation steps before the accuracy numbers can be taken at face value.

Referee Report

2 major / 1 minor

Summary. The paper introduces Decisive, an interactive decision-making framework that extracts an option-scoring matrix from unstructured documents to ground decisions objectively, then uses Bayesian inference to learn a user's low-dimensional latent preference vector via adaptively selected pairwise tradeoff questions that maximize information gain, claiming up to 20% gains in decision accuracy over general-purpose LLMs and existing frameworks across domains.

Significance. If the document extraction step can be shown to be reliable and the experimental results hold under scrutiny, the work could advance interactive decision support by combining document grounding with efficient, transparent preference elicitation that minimizes user effort. The information-gain selection mechanism is a standard but well-applied strength here.

major comments (2)

[Abstract] Abstract: the central claim of 'up to 20% improvement in decision accuracy over strong baselines' is stated without any details on experimental design, baseline implementations, statistical tests, error bars, or domain specifications, making it impossible to assess whether the data support the claim.
[Method] Method section (option-scoring matrix extraction): the framework's performance claims rest on the assumption that an accurate and complete option-scoring matrix can be reliably extracted from unstructured documents, yet no human-validated extraction accuracy, inter-annotator agreement, or ablation on extraction noise is reported; this is load-bearing because any downstream Bayesian gains cannot be attributed to the elicitation if the matrix is systematically distorted.

minor comments (1)

[Abstract] The abstract could clarify the specific domains tested and the exact definition of 'decision accuracy' used in the experiments.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for their insightful comments on our paper. We provide point-by-point responses to the major comments and indicate the revisions we will make to the manuscript.

read point-by-point responses

Referee: [Abstract] Abstract: the central claim of 'up to 20% improvement in decision accuracy over strong baselines' is stated without any details on experimental design, baseline implementations, statistical tests, error bars, or domain specifications, making it impossible to assess whether the data support the claim.

Authors: We agree that the abstract would be strengthened by including more details about the claim. In the revised manuscript, we have updated the abstract to mention the specific domains (product selection, travel planning, and medical decision support), the baselines (general-purpose LLMs such as GPT-4 and existing decision frameworks), and that the improvements are statistically significant with error bars provided in the main experimental results. This allows readers to better evaluate the claim without exceeding abstract length limits. revision: yes
Referee: [Method] Method section (option-scoring matrix extraction): the framework's performance claims rest on the assumption that an accurate and complete option-scoring matrix can be reliably extracted from unstructured documents, yet no human-validated extraction accuracy, inter-annotator agreement, or ablation on extraction noise is reported; this is load-bearing because any downstream Bayesian gains cannot be attributed to the elicitation if the matrix is systematically distorted.

Authors: This comment correctly identifies a gap in the original submission. We did not report human validation or noise ablation for the matrix extraction. We will add an ablation study in the revised Experiments section that simulates extraction noise by perturbing the scoring matrix entries and measures the resulting change in decision accuracy. This will show that the adaptive elicitation provides benefits even under moderate distortion. However, a full human validation study with inter-annotator agreement was not conducted in the original work and would require additional resources; we note this as a limitation and plan to address it in follow-up research. revision: partial

standing simulated objections not resolved

Full human-validated extraction accuracy and inter-annotator agreement for the option-scoring matrix extraction, as this was not performed in the original experiments.

Circularity Check

0 steps flagged

No circularity: standard Bayesian elicitation loop with independent extraction step

full rationale

The paper describes a pipeline that first extracts an option-scoring matrix from documents (treated as an external input) and then runs a standard Bayesian update on a latent preference vector using information-gain-selected pairwise queries. No equation, algorithm, or claim in the abstract or described framework reduces the final accuracy improvement or the preference vector itself to a fitted parameter or self-referential quantity by construction. The derivation chain remains open to external validation of the extraction accuracy and does not collapse into its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Based solely on the abstract, the central claim rests on standard assumptions from Bayesian inference and information theory plus the domain assumption that document-to-matrix extraction is feasible; no explicit free parameters, new entities, or ad-hoc axioms are named.

axioms (2)

domain assumption User preferences can be represented as a latent vector that is updated via Bayesian inference from pairwise comparisons
Invoked in the description of active learning of the preference vector
domain assumption An objective option-scoring matrix can be extracted from unstructured source documents
Stated as the grounding step for decisions

pith-pipeline@v0.9.0 · 5471 in / 1348 out tokens · 47336 ms · 2026-05-10T05:08:14.098789+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

300 extracted references · 129 canonical work pages

[1]

Scaling Learning Algorithms Towards

Bengio, Yoshua and LeCun, Yann , booktitle =. Scaling Learning Algorithms Towards
[2]

and Osindero, Simon and Teh, Yee Whye , journal =

Hinton, Geoffrey E. and Osindero, Simon and Teh, Yee Whye , journal =. A Fast Learning Algorithm for Deep Belief Nets , volume =
[3]

2016 , publisher=

Deep learning , author=. 2016 , publisher=

2016
[4]

2025 , eprint=

Exploring the Impact of Explainable AI and Cognitive Capabilities on Users' Decisions , author=. 2025 , eprint=

2025
[5]

Perspectives on Psychological Science , year =

Mark Steyvers and Akshay Kumar , title =. Perspectives on Psychological Science , year =. doi:10.1177/17456916231181102 , pmid =

work page doi:10.1177/17456916231181102
[6]

2024 , eprint=

DeLLMa: Decision Making Under Uncertainty with Large Language Models , author=. 2024 , eprint=

2024
[7]

2025 , eprint=

DecisionFlow: Advancing Large Language Model as Principled Decision Maker , author=. 2025 , eprint=

2025
[8]

Jurnal Informatika Ekonomi Bisnis , author=

Artificial Intelligence in Financial Decision-Making: Opportunities and Challenges for Investment Strategies , volume=. Jurnal Informatika Ekonomi Bisnis , author=. 2025 , month=. doi:10.37034/infeb.v7i2.1140 , abstractNote=

work page doi:10.37034/infeb.v7i2.1140 2025
[9]

Education Sciences , VOLUME =

Shi, Lehong and Ding, Ai-Chu (Elisha) and Choi, Ikseon , TITLE =. Education Sciences , VOLUME =. 2024 , NUMBER =

2024
[10]

Health Services Research and Managerial Epidemiology , volume =

Khosravi, Maryam and Zare, Zahra and Mojtabaeian, Seyed Mohammad and Izadi, Reza , title =. Health Services Research and Managerial Epidemiology , volume =. 2024 , publisher =

2024
[11]

2024 , eprint=

Language Models are Alignable Decision-Makers: Dataset and Application to the Medical Triage Domain , author=. 2024 , eprint=

2024
[12]

2024 , eprint=

Large language models in healthcare and medical domain: A review , author=. 2024 , eprint=

2024
[13]

Large language models-powered clinical decision support: enhancing or replacing human expertise? , journal =

Jia Li and Zichun Zhou and Han Lyu and Zhenchang Wang , keywords =. Large language models-powered clinical decision support: enhancing or replacing human expertise? , journal =. 2025 , issn =. doi:https://doi.org/10.1016/j.imed.2025.01.001 , url =

work page doi:10.1016/j.imed.2025.01.001 2025
[14]

& Saikia, M

Maity, S. and Saikia, M. J. , title =. Bioengineering (Basel) , year =. doi:10.3390/bioengineering12060631 , pmid =

work page doi:10.3390/bioengineering12060631
[15]

2024 , eprint=

MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making , author=. 2024 , eprint=

2024
[16]

Indraprasta and Ach Salman Faridzi and Bagus M

Khanisyah Erza Gumilar and Birama R. Indraprasta and Ach Salman Faridzi and Bagus M. Wibowo and Aditya Herlambang and Eccita Rahestyningtyas and Budi Irawan and Zulkarnain Tambunan and Ahmad Fadhli Bustomi and Bagus Ngurah Brahmantara and Zih-Ying Yu and Yu-Cheng Hsu and Herlangga Pramuditya and Very Great E. Putra and Hari Nugroho and Pungky Mulawardhana...

work page doi:10.1016/j.csbj.2024.10.050 2024
[17]

Genay, M

Foo, Charisse and Foong, Pin Sym and Nadal, Camille and Ureyang, Natasha and Naylin, Thant and Koh, Gerald Choon Huat , title =. Proceedings of the 2025 ACM Designing Interactive Systems Conference , pages =. 2025 , isbn =. doi:10.1145/3715336.3735779 , abstract =

work page doi:10.1145/3715336.3735779 2025
[18]

2024 , eprint=

Determinants of LLM-assisted Decision-Making , author=. 2024 , eprint=

2024
[19]

Management Journal for Advanced Research , year =

Alexander Changeux and Stephen Montagnier , title =. Management Journal for Advanced Research , year =. doi:10.5281/zenodo.13444483 , url =

work page doi:10.5281/zenodo.13444483
[20]

2025 , eprint=

Large Language Models for Supply Chain Decisions , author=. 2025 , eprint=

2025
[21]

2024 , eprint=

INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent , author=. 2024 , eprint=

2024
[22]

2024 , eprint=

FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making , author=. 2024 , eprint=

2024
[23]

2024 , eprint=

MAPLE: A Framework for Active Preference Learning Guided by Large Language Models , author=. 2024 , eprint=

2024
[24]

The First Workshop on System-2 Reasoning at Scale, NeurIPS'24 , year=

Can Language Models Perform Implicit Bayesian Inference Over User Preference States? , author=. The First Workshop on System-2 Reasoning at Scale, NeurIPS'24 , year=
[25]

2024 , eprint=

STaR-GATE: Teaching Language Models to Ask Clarifying Questions , author=. 2024 , eprint=

2024
[26]

2025 , eprint=

TO-GATE: Clarifying Questions and Summarizing Responses with Trajectory Optimization for Eliciting Human Preference , author=. 2025 , eprint=

2025
[27]

2025 , eprint=

Modeling Future Conversation Turns to Teach LLMs to Ask Clarifying Questions , author=. 2025 , eprint=

2025
[28]

2024 , eprint=

Active Preference Inference using Language Models and Probabilistic Reasoning , author=. 2024 , eprint=

2024
[29]

Language Models are Alignable Decision-Makers: Dataset and Application to the Medical Triage Domain

Hu, Brian and Ray, Bill and Leung, Alice and Summerville, Amy and Joy, David and Funk, Christopher and Basharat, Arslan. Language Models are Alignable Decision-Makers: Dataset and Application to the Medical Triage Domain. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Techn...

work page doi:10.18653/v1/2024.naacl-industry.18 2024
[30]

2025 , eprint=

ChoiceMates: Supporting Unfamiliar Online Decision-Making with Multi-Agent Conversational Interactions , author=. 2025 , eprint=

2025
[31]

2024 , eprint=

Decision-Making Behavior Evaluation Framework for LLMs under Uncertain Context , author=. 2024 , eprint=

2024
[32]

Clarify When Necessary: Resolving Ambiguity Through Interaction with LM s

Zhang, Michael JQ and Choi, Eunsol. Clarify When Necessary: Resolving Ambiguity Through Interaction with LM s. Findings of the Association for Computational Linguistics: NAACL 2025. 2025. doi:10.18653/v1/2025.findings-naacl.306

work page doi:10.18653/v1/2025.findings-naacl.306 2025
[33]

Teaching

Chang, Shane , month = may, year =. Teaching
[34]

2025 , eprint=

InfoQuest: Evaluating Multi-Turn Dialogue Agents for Open-Ended Conversations with Hidden Context , author=. 2025 , eprint=

2025
[35]

2025 , eprint=

Do LLMs Recognize Your Latent Preferences? A Benchmark for Latent Information Discovery in Personalized Interaction , author=. 2025 , eprint=

2025
[36]

2025 , eprint=

Asking Clarifying Questions for Preference Elicitation With Large Language Models , author=. 2025 , eprint=

2025
[37]

2024 , eprint=

Bayesian Preference Elicitation with Language Models , author=. 2024 , eprint=

2024
[38]

Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems , articleno =

Ma, Shuai and Chen, Qiaoyi and Wang, Xinru and Zheng, Chengbo and Peng, Zhenhui and Yin, Ming and Ma, Xiaojuan , title =. Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems , articleno =. 2025 , isbn =. doi:10.1145/3706598.3713423 , abstract =

work page doi:10.1145/3706598.3713423 2025
[39]

Generative AI for decision-making: A multidisciplinary perspective , journal =

Mousa Albashrawi , keywords =. Generative AI for decision-making: A multidisciplinary perspective , journal =. 2025 , issn =. doi:https://doi.org/10.1016/j.jik.2025.100751 , url =

work page doi:10.1016/j.jik.2025.100751 2025
[40]

Scientific Reports , author =

Responsible. Scientific Reports , author =. 2025 , pages =. doi:10.1038/s41598-025-22127-7 , language =

work page doi:10.1038/s41598-025-22127-7 2025
[41]

J ob F air: A Framework for Benchmarking Gender Hiring Bias in Large Language Models

Wang, Ze and Wu, Zekun and Guan, Xin and Thaler, Michael and Koshiyama, Adriano and Lu, Skylar and Beepath, Sachin and Ertekin, Ediz and Perez-Ortiz, Maria. J ob F air: A Framework for Benchmarking Gender Hiring Bias in Large Language Models. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024. doi:10.18653/v1/2024.findings-emnlp.184

work page doi:10.18653/v1/2024.findings-emnlp.184 2024
[42]

2023 , eprint=

Are Emily and Greg Still More Employable than Lakisha and Jamal? Investigating Algorithmic Hiring Bias in the Era of ChatGPT , author=. 2023 , eprint=

2023
[43]

2021 , url =

Snehaan Bhawal , title =. 2021 , url =

2021
[44]

2025 , eprint=

Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs , author=. 2025 , eprint=

2025
[45]

2024 , eprint=

UMBRELA: UMbrela is the (Open-Source Reproduction of the) Bing RELevance Assessor , author=. 2024 , eprint=

2024
[46]

Doucet, Arnaud and Johansen, Adam M. , url =. The Oxford handbook of nonlinear filtering , title =. 2011 , series =

2011
[47]

IEEE Trans

A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , author=. IEEE Trans. Signal Process. , year=
[48]

Pu, Pearl and Chen, Li , title =. AI Mag. , month = dec, pages =. 2008 , issue_date =. doi:10.1609/aimag.v29i4.2200 , abstract =

work page doi:10.1609/aimag.v29i4.2200 2008
[49]

Proceedings of the 6th ACM Conference on Electronic Commerce , pages =

Pu, Pearl and Chen, Li , title =. Proceedings of the 6th ACM Conference on Electronic Commerce , pages =. 2005 , isbn =. doi:10.1145/1064009.1064038 , abstract =

work page doi:10.1145/1064009.1064038 2005
[50]

Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics , pages =

Real-time Multiattribute Bayesian Preference Elicitation with Pairwise Comparison Queries , author =. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics , pages =. 2010 , editor =

2010
[51]

Proceedings of the 18th ACM Conference on Recommender Systems , pages =

Austin, David and Korikov, Anton and Toroghi, Armin and Sanner, Scott , title =. Proceedings of the 18th ACM Conference on Recommender Systems , pages =. 2024 , isbn =. doi:10.1145/3640457.3688142 , abstract =

work page doi:10.1145/3640457.3688142 2024
[52]

Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022). 2022

2022
[53]

A Systematic Survey of Text Worlds as Embodied Natural Language Environments

Jansen, Peter. A Systematic Survey of Text Worlds as Embodied Natural Language Environments. Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022). 2022. doi:10.18653/v1/2022.wordplay-1.1

work page doi:10.18653/v1/2022.wordplay-1.1 2022
[54]

A Minimal Computational Improviser Based on Oral Thought

Montfort, Nick and Bartlett Fernandez, Sebastian. A Minimal Computational Improviser Based on Oral Thought. Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022). 2022. doi:10.18653/v1/2022.wordplay-1.2

work page doi:10.18653/v1/2022.wordplay-1.2 2022
[55]

Craft an Iron Sword: Dynamically Generating Interactive Game Characters by Prompting Large Language Models Tuned on Code

Volum, Ryan and Rao, Sudha and Xu, Michael and DesGarennes, Gabriel and Brockett, Chris and Van Durme, Benjamin and Deng, Olivia and Malhotra, Akanksha and Dolan, Bill. Craft an Iron Sword: Dynamically Generating Interactive Game Characters by Prompting Large Language Models Tuned on Code. Proceedings of the 3rd Wordplay: When Language Meets Games Worksho...

work page doi:10.18653/v1/2022.wordplay-1.3 2022
[56]

A Sequence Modelling Approach to Question Answering in Text-Based Games

Furman, Gregory and Toledo, Edan and Shock, Jonathan and Buys, Jan. A Sequence Modelling Approach to Question Answering in Text-Based Games. Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022). 2022. doi:10.18653/v1/2022.wordplay-1.4

work page doi:10.18653/v1/2022.wordplay-1.4 2022
[57]

Automatic Exploration of Textual Environments with Language-Conditioned Autotelic Agents

Teodorescu, Laetitia and Yuan, Xingdi and C \^o t \'e , Marc-Alexandre and Oudeyer, Pierre-Yves. Automatic Exploration of Textual Environments with Language-Conditioned Autotelic Agents. Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022). 2022. doi:10.18653/v1/2022.wordplay-1.5

work page doi:10.18653/v1/2022.wordplay-1.5 2022
[58]

Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022

2022
[59]

Separating Hate Speech and Offensive Language Classes via Adversarial Debiasing

Yuan, Shuzhou and Maronikolakis, Antonis and Sch. Separating Hate Speech and Offensive Language Classes via Adversarial Debiasing. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.1

work page doi:10.18653/v1/2022.woah-1.1 2022
[60]

Towards Automatic Generation of Messages Countering Online Hate Speech and Microaggressions

Ashida, Mana and Komachi, Mamoru. Towards Automatic Generation of Messages Countering Online Hate Speech and Microaggressions. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.2

work page doi:10.18653/v1/2022.woah-1.2 2022
[61]

G rease V ision: Rewriting the Rules of the Interface

Datta, Siddhartha and Kollnig, Konrad and Shadbolt, Nigel. G rease V ision: Rewriting the Rules of the Interface. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.3

work page doi:10.18653/v1/2022.woah-1.3 2022
[62]

Improving Generalization of Hate Speech Detection Systems to Novel Target Groups via Domain Adaptation

Ludwig, Florian and Dolos, Klara and Zesch, Torsten and Hobley, Eleanor. Improving Generalization of Hate Speech Detection Systems to Novel Target Groups via Domain Adaptation. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.4

work page doi:10.18653/v1/2022.woah-1.4 2022
[63]

`` Zo Grof ! '' : A Comprehensive Corpus for Offensive and Abusive Language in D utch

Ruitenbeek, Ward and Zwart, Victor and Van Der Noord, Robin and Gnezdilov, Zhenja and Caselli, Tommaso. `` Zo Grof ! '' : A Comprehensive Corpus for Offensive and Abusive Language in D utch. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.5

work page doi:10.18653/v1/2022.woah-1.5 2022
[64]

Counter- TWIT : An I talian Corpus for Online Counterspeech in Ecological Contexts

Goffredo, Pierpaolo and Basile, Valerio and Cepollaro, Bianca and Patti, Viviana. Counter- TWIT : An I talian Corpus for Online Counterspeech in Ecological Contexts. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.6

work page doi:10.18653/v1/2022.woah-1.6 2022
[65]

S tereo KG : Data-Driven Knowledge Graph Construction For Cultural Knowledge and Stereotypes

Deshpande, Awantee and Ruiter, Dana and Mosbach, Marius and Klakow, Dietrich. S tereo KG : Data-Driven Knowledge Graph Construction For Cultural Knowledge and Stereotypes. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.7

work page doi:10.18653/v1/2022.woah-1.7 2022
[66]

The subtle language of exclusion: Identifying the Toxic Speech of Trans-exclusionary Radical Feminists

Lu, Christina and Jurgens, David. The subtle language of exclusion: Identifying the Toxic Speech of Trans-exclusionary Radical Feminists. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.8

work page doi:10.18653/v1/2022.woah-1.8 2022
[67]

Lost in Distillation: A Case Study in Toxicity Modeling

Chvasta, Alyssa and Lees, Alyssa and Sorensen, Jeffrey and Vasserman, Lucy and Goyal, Nitesh. Lost in Distillation: A Case Study in Toxicity Modeling. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.9

work page doi:10.18653/v1/2022.woah-1.9 2022
[68]

Cleansing & expanding the HURTLEX (el) with a multidimensional categorization of offensive words

Stamou, Vivian and Alexiou, Iakovi and Klimi, Antigone and Molou, Eleftheria and Saivanidou, Alexandra and Markantonatou, Stella. Cleansing & expanding the HURTLEX (el) with a multidimensional categorization of offensive words. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.10

work page doi:10.18653/v1/2022.woah-1.10 2022
[69]

Free speech or Free Hate Speech? Analyzing the Proliferation of Hate Speech in Parler

Israeli, Abraham and Tsur, Oren. Free speech or Free Hate Speech? Analyzing the Proliferation of Hate Speech in Parler. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.11

work page doi:10.18653/v1/2022.woah-1.11 2022
[70]

Resources for Multilingual Hate Speech Detection

Arango Monnar, Ayme and Perez, Jorge and Poblete, Barbara and Salda \ n a, Magdalena and Proust, Valentina. Resources for Multilingual Hate Speech Detection. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.12

work page doi:10.18653/v1/2022.woah-1.12 2022
[71]

Enriching Abusive Language Detection with Community Context

Saleem, Haji Mohammad and Kurrek, Jana and Ruths, Derek. Enriching Abusive Language Detection with Community Context. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.13

work page doi:10.18653/v1/2022.woah-1.13 2022
[72]

DeTox: A Comprehensive Dataset for G erman Offensive Language and Conversation Analysis

Demus, Christoph and Pitz, Jonas and Sch. DeTox: A Comprehensive Dataset for G erman Offensive Language and Conversation Analysis. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.14

work page doi:10.18653/v1/2022.woah-1.14 2022
[73]

Multilingual H ate C heck: Functional Tests for Multilingual Hate Speech Detection Models

R. Multilingual H ate C heck: Functional Tests for Multilingual Hate Speech Detection Models. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.15

work page doi:10.18653/v1/2022.woah-1.15 2022
[74]

Distributional properties of political dogwhistle representations in S wedish BERT

Hertzberg, Niclas and Cooper, Robin and Lindgren, Elina and R. Distributional properties of political dogwhistle representations in S wedish BERT. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.16

work page doi:10.18653/v1/2022.woah-1.16 2022
[75]

Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions

Khurana, Urja and Vermeulen, Ivar and Nalisnick, Eric and Van Noorloos, Marloes and Fokkens, Antske. Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.17

work page doi:10.18653/v1/2022.woah-1.17 2022
[76]

Accounting for Offensive Speech as a Practice of Resistance

Diaz, Mark and Amironesei, Razvan and Weidinger, Laura and Gabriel, Iason. Accounting for Offensive Speech as a Practice of Resistance. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.18

work page doi:10.18653/v1/2022.woah-1.18 2022
[77]

Towards a Multi-Entity Aspect-Based Sentiment Analysis for Characterizing Directed Social Regard in Online Messaging

Zheng, Joan and Friedman, Scott and Schmer-galunder, Sonja and Magnusson, Ian and Wheelock, Ruta and Gottlieb, Jeremy and Gomez, Diana and Miller, Christopher. Towards a Multi-Entity Aspect-Based Sentiment Analysis for Characterizing Directed Social Regard in Online Messaging. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:1...

work page doi:10.18653/v1/2022.woah-1.19 2022
[78]

Flexible text generation for counterfactual fairness probing

Fryer, Zee and Axelrod, Vera and Packer, Ben and Beutel, Alex and Chen, Jilin and Webster, Kellie. Flexible text generation for counterfactual fairness probing. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.20

work page doi:10.18653/v1/2022.woah-1.20 2022
[79]

Users Hate Blondes: Detecting Sexism in User Comments on Online R omanian News

Moldovan, Andreea and Cs. Users Hate Blondes: Detecting Sexism in User Comments on Online R omanian News. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.21

work page doi:10.18653/v1/2022.woah-1.21 2022
[80]

Targeted Identity Group Prediction in Hate Speech Corpora

Sachdeva, Pratik and Barreto, Renata and Von Vacano, Claudia and Kennedy, Chris. Targeted Identity Group Prediction in Hate Speech Corpora. Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). 2022. doi:10.18653/v1/2022.woah-1.22

work page doi:10.18653/v1/2022.woah-1.22 2022

Showing first 80 references.