1 citations · 1 across the 13 of their papers we have counts for
13 papers
WOLF: Werewolf-based Observations for LLM Deception and Falsehoods
Mrinal Agarwal, Saad Rana, Theo Sundoro +5
Deception is a fundamental challenge for multi-agent reasoning: effective systems must strategically conceal information while detecting misleading behavior in others. Yet most eva…
Interpreting the Latent Structure of Operator Precedence in Language Models
Dharunish Yugeswardeenoo, Harshil Nukala, Ved Shah +4
Large Language Models (LLMs) have demonstrated impressive reasoning capabilities but continue to struggle with arithmetic tasks. Prior works largely focus on outputs or prompting s…
ERGO: Entropy-guided Resetting for Generation Optimization in Multi-turn Language Models
Haziq Mohammad Khalid, Athikash Jeyaganthan, Timothy Do +4
Large Language Models (LLMs) suffer significant performance degradation in multi-turn conversations when information is presented incrementally. Given that multi-turn conversations…
Adaptive Originality Filtering: Rejection Based Prompting and RiddleScore for Culturally Grounded Multilingual Riddle Generation
Duy Le, Kent Ziti, Evan Girard-Sun +4
Language models are increasingly tested on multilingual creativity, demanding culturally grounded, abstract generations. Standard prompting methods often produce repetitive or shal…
Universal Neurons in GPT-2: Emergence, Persistence, and Functional Impact
Advey Nandan, Cheng-Ting Chou, Amrit Kurakula +4
We investigate the phenomenon of neuron universality in independently trained GPT-2 Small models, examining these universal neurons-neurons with consistently correlated activations…
The Geometry of Harmfulness in LLMs through Subconcept Probing
McNair Shah, Saleena Angeline, Adhitya Rajendra Kumar +5
Recent advances in large language models (LLMs) have intensified the need to understand and reliably curb their harmful behaviours. We introduce a multidimensional framework for pr…
Semantic Convergence: Investigating Shared Representations Across Scaled LLMs
Daniel Son, Sanjana Rathore, Andrew Rufail +6
We investigate feature universality in Gemma-2 language models (Gemma-2-2B and Gemma-2-9B), asking whether models with a four-fold difference in scale still converge on comparable…
COREVQA: A Crowd Observation and Reasoning Entailment Visual Question Answering Benchmark
Ishant Chintapatla, Kazuma Choji, Naaisha Agarwal +6
Recently, many benchmarks and datasets have been developed to evaluate Vision-Language Models (VLMs) using visual question answering (VQA) pairs, and models have shown significant…
Rewrite-to-Rank: Optimizing Ad Visibility via Retrieval-Aware Text Rewriting
Chloe Ho, Ishneet Sukhvinder Singh, Diya Sharma +4
Search algorithms and user query relevance have given LLMs the ability to return relevant information, but the effect of content phrasing on ad visibility remains underexplored. We…
Causal Language Control in Multilingual Transformers via Sparse Feature Steering
Cheng-Ting Chou, George Liu, Jessica Sun +4
Deterministically controlling the target generation language of large multilingual language models (LLMs) remains a fundamental challenge, particularly in zero-shot settings where…
Distill CLIP (DCLIP): Enhancing Image-Text Retrieval via Cross-Modal Transformer Distillation
Daniel Csizmadia, Andrei Codreanu, Victor Sim +5
We present Distill CLIP (DCLIP), a fine-tuned variant of the CLIP model that enhances multimodal image-text retrieval while preserving the original model's strong zero-shot classif…
From Directions to Cones: Exploring Multidimensional Representations of Propositional Facts in LLMs
Stanley Yu, Vaidehi Bulusu, Oscar Yasunaga +5
Large Language Models (LLMs) exhibit strong conversational abilities but often generate falsehoods. Prior work suggests that the truthfulness of simple propositions can be represen…
Rosetta-PL: Propositional Logic as a Benchmark for Large Language Model Reasoning
Shaun Baek, Shaun Esua-Mensah, Cyrus Tsui +6
Large Language Models (LLMs) are primarily trained on high-resource natural languages, limiting their effectiveness in low-resource settings and in tasks requiring deep logical rea…