Alert button
Picture for Mert Yuksekgonul

Mert Yuksekgonul

Alert button

How Well Can LLMs Negotiate? NegotiationArena Platform and Analysis

Feb 08, 2024
Federico Bianchi, Patrick John Chia, Mert Yuksekgonul, Jacopo Tagliabue, Dan Jurafsky, James Zou

Viaarxiv icon

ChatGPT Exhibits Gender and Racial Biases in Acute Coronary Syndrome Management

Nov 10, 2023
Angela Zhang, Mert Yuksekgonul, Joshua Guild, James Zou, Joseph C. Wu

Viaarxiv icon

KITAB: Evaluating LLMs on Constraint Satisfaction for Information Retrieval

Oct 24, 2023
Marah I Abdin, Suriya Gunasekar, Varun Chandrasekaran, Jerry Li, Mert Yuksekgonul, Rahee Ghosh Peshawaria, Ranjita Naik, Besmira Nushi

Viaarxiv icon

Diversity of Thought Improves Reasoning Abilities of Large Language Models

Oct 11, 2023
Ranjita Naik, Varun Chandrasekaran, Mert Yuksekgonul, Hamid Palangi, Besmira Nushi

Figure 1 for Diversity of Thought Improves Reasoning Abilities of Large Language Models
Figure 2 for Diversity of Thought Improves Reasoning Abilities of Large Language Models
Figure 3 for Diversity of Thought Improves Reasoning Abilities of Large Language Models
Figure 4 for Diversity of Thought Improves Reasoning Abilities of Large Language Models
Viaarxiv icon

Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models

Sep 26, 2023
Mert Yuksekgonul, Varun Chandrasekaran, Erik Jones, Suriya Gunasekar, Ranjita Naik, Hamid Palangi, Ece Kamar, Besmira Nushi

Figure 1 for Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models
Figure 2 for Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models
Figure 3 for Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models
Figure 4 for Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models
Viaarxiv icon

Beyond Confidence: Reliable Models Should Also Consider Atypicality

May 29, 2023
Mert Yuksekgonul, Linjun Zhang, James Zou, Carlos Guestrin

Figure 1 for Beyond Confidence: Reliable Models Should Also Consider Atypicality
Figure 2 for Beyond Confidence: Reliable Models Should Also Consider Atypicality
Figure 3 for Beyond Confidence: Reliable Models Should Also Consider Atypicality
Figure 4 for Beyond Confidence: Reliable Models Should Also Consider Atypicality
Viaarxiv icon

Discover and Cure: Concept-aware Mitigation of Spurious Correlation

May 01, 2023
Shirley Wu, Mert Yuksekgonul, Linjun Zhang, James Zou

Figure 1 for Discover and Cure: Concept-aware Mitigation of Spurious Correlation
Figure 2 for Discover and Cure: Concept-aware Mitigation of Spurious Correlation
Figure 3 for Discover and Cure: Concept-aware Mitigation of Spurious Correlation
Figure 4 for Discover and Cure: Concept-aware Mitigation of Spurious Correlation
Viaarxiv icon

GPT detectors are biased against non-native English writers

Apr 18, 2023
Weixin Liang, Mert Yuksekgonul, Yining Mao, Eric Wu, James Zou

Figure 1 for GPT detectors are biased against non-native English writers
Figure 2 for GPT detectors are biased against non-native English writers
Viaarxiv icon

SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysis

Feb 01, 2023
Roxana Daneshjou, Mert Yuksekgonul, Zhuo Ran Cai, Roberto Novoa, James Zou

Figure 1 for SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysis
Figure 2 for SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysis
Figure 3 for SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysis
Figure 4 for SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysis
Viaarxiv icon