3.5k citations · 14.1k across the 436 of their papers we have counts for
112 papers · 1 filter
How Much Do Activation Monitors Benefit from Original Model Computations?
James Oldfield, Shawn Im, Jay J Culligan +3
One powerful way to safeguard AI systems is to monitor their internal computations--where signs of undesirable behavior might be more easily detectable than from models' inputs and…
Inside the Loop: A Mechanistic Study of Weight-Tied Transformers on Depth-Bound Algorithmic Tasks
Magnus Sesodia, Juil Sock, Philip Torr +1
Solving algorithmic tasks with many sequential steps typically requires scaling up transformer depth or chain-of-thought length, incurring unfavorable parameter and inference-compu…
On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D
Fabio J. Fehr, Philip Torr
General-purpose models can adapt to many tasks from context, while specialised models can execute individual functions with less capacity. Yet obtaining such specialists requires t…
Towards Better Exploration in Sequential Test-Time Scaling
Joseph Rance, Fabio Pizzati, Juil Sock +4
Test-time scaling improves language model reasoning by spending additional compute at inference. However, both classes of existing methods often fail to continue improving over lon…
Procedural Pretraining for Molecular Property Prediction
Moritz Friedemann, Zachary Shinnick, Philip Torr +1
Molecular property prediction is often limited by the small size of labeled downstream datasets, motivating pretraining on large corpora of unlabeled molecules. In this work, we as…
Online Surrogate Repair: Decoupling High-Fidelity Feedback from Search Length in Closed-Loop Discovery
Xiaotang Feng, Philip Torr, Bruno Andreis
Closed-loop AI scientists can generate candidate designs at low marginal computational cost, whereas reliable feedback may require wet-lab synthesis, characterization, or high-fide…
Conformalized Large Language Models under Configuration Shift
Yuqicheng Zhu, Jialin Yu, Lin Li +7
Conformal prediction (CP) is a distribution-free framework for uncertainty quantification that has recently been adapted to large language models (LLMs), providing prediction sets…
A Low-Rank Subspace Analysis of LLM Interventions
Angira Sharma, Christian Schroeder de Witt, Philip Torr +2
Interventions designed to modify a particular behavior in LLMs, such as refusal or sycophancy, often produce unintended changes in other behaviors. This lack of targeted control ma…
When Language Representations Interact: Separability and Cross-Lingual Effects in LLMs
Boris Marinov, Angira Sharma, Christian Schroeder de Witt +3
Large language models exhibit strong multilingual capabilities, however, their internal representations are difficult to interpret. Understanding these interactions is important fo…
Running the Gauntlet: Hard Agentic Tasks
Mykola Vysotskyi, Runqi Lin, Grzegorz Biziel +21
As agentic systems continue to evolve and are widely deployed in real-world scenarios, there is a growing demand to faithfully evaluate their capabilities. However, current benchma…
Dynamic Mixture of Latent Memories for Self-Evolving Agents
Dianzhi Yu, Vireo Zhang, Hongru Wang +7
Achieving self-evolution in intelligent agents requires the continual accumulation of new knowledge across changing task sequences without forgetting previously acquired abilities.…
Distinguishable Deletion: Unifying Knowledge Erasure and Refusal for Large Language Model Unlearning
Puning Yang, Junchi Yu, Qizhou Wang +3
Mitigating sensitive and harmful outputs is fundamental to ensuring safe deployment of LLMs. Existing approaches typically follow two paradigms: Knowledge Deletion (KD), which eras…
Contrastive Image-Metadata Pre-Training for Materials Transmission Electron Microscopy
Georgia Channing, Debora Keller, Marta D. Rossell +3
The transmission electron microscope facilitates the highest-resolution imaging of any instrument ever created, and its limiting factor is no longer spatial resolution but dose eff…
LongCoT: Benchmarking Long-Horizon Chain-of-Thought Reasoning
Sumeet Ramesh Motwani, Daniel Nichols, Charles London +17
As language models are increasingly deployed for complex autonomous tasks, their ability to reason accurately over longer horizons becomes critical. An essential component of this…
Improving Semantic Uncertainty Quantification in Language Model Question-Answering via Token-Level Temperature Scaling
Tom A. Lamb, Desi R. Ivanova, Philip H. S. Torr +1
Calibration is central to reliable semantic uncertainty quantification, yet prior work has largely focused on discrimination, neglecting calibration. As calibration and discriminat…
Safer by Diffusion, Broken by Context: Diffusion LLM's Safety Blessing and Its Failure Mode
Zeyuan He, Yupeng Chen, Lang Lin +7
Diffusion large language models (D-LLMs) offer an alternative to autoregressive LLMs (AR-LLMs) and have demonstrated advantages in generation efficiency. Beyond the utility benefit…
AutoOR: Scalably Post-training LLMs to Autoformulate Operations Research Problems
Sumeet Ramesh Motwani, Chuan Du, Aleksander Petrov +4
Optimization problems are central to decision-making in manufacturing, logistics, scheduling, and other industrial settings. Translating complicated descriptions of these problems…
Measuring Progress in Reasoning Toward Mathematical Discovery with Automatic Verification
Erik Y. Wang, Sumeet R. Motwani, James V. Roggeveen +9
Can AI make progress on important, unsolved mathematical problems? Large language models are now capable of sophisticated mathematical and scientific reasoning, but whether they ca…
MechPert: Mechanistic Consensus as an Inductive Bias for Unseen Perturbation Prediction
Marc Boubnovski Martell, Josefa Lia Stoisser, Lawrence Phillips +6
Predicting transcriptional responses to unseen genetic perturbations is essential for understanding gene regulation and prioritizing large-scale perturbation experiments. Existing…
Gradient Residual Connections
Yangchen Pan, Qizhen Ying, Philip Torr +1
Existing work has linked properties of a function's gradient to the difficulty of function approximation. Motivated by these insights, we study how gradient information can be leve…
Reliable and Responsible Foundation Models: A Comprehensive Survey
Xinyu Yang, Junlin Han, Rishi Bommasani +49
Foundation models, including Large Language Models (LLMs), Multimodal Large Language Models (MLLMs), Image Generative Models (i.e, Text-to-Image Models and Image-Editing Models), a…
The Alignment Curse: Modality Alignment Supercharges Audio Attacks via Text Transfer
Yupeng Chen, Junchi Yu, Aoxi Liu +3
Recent advances in end-to-end trained omni-models have substantially improved audio capabilities by strengthening text-audio modality alignment. However, whether such alignment ina…
Too Late to Recall: Explaining the Two-Hop Problem in Multimodal Knowledge Retrieval
Constantin Venhoff, Ashkan Khakzar, Sonia Joseph +2
Training vision language models (VLMs) aims to align visual representations from a vision encoder with the textual representations of a pretrained large language model (LLM). Howev…
DEEDEE: Fast and Scalable Out-of-Distribution Dynamics Detection
Tala Aljaafari, Varun Kanade, Philip Torr +1
Deploying reinforcement learning (RL) in safety-critical settings is constrained by brittleness under distribution shift. We study out-of-distribution (OOD) detection for RL time s…