2 citations · 2 across the 9 of their papers we have counts for
7 papers · 1 filter
Test-Time Adaptation of Reasoning Strategies with Bayesian Nonparametric Memory
Keshav Ramji, Tahira Naseem, Ramón Fernandez Astudillo
While modern large language models (LLMs) have been trained to reason through verbalized chains-of-thought, the generation cost grows substantially due to suboptimal paths to reach…
The Surprising Effectiveness of Shared Memory in Looped Transformers
Giovanni Monea, Keshav Ramji, Yousef El-Kurdi +4
Looped Transformers apply the same layers several times per token, adding compute to improve quality without more parameters. Each recursion, however, writes its own key-value cach…
Learning to Predict Distributions over Weight Updates for Test-Time Adaptation
Azal Ahmad Khan, Keshav Ramji, Tahira Naseem +2
Hypernetworks have recently shown success in dynamically adapting the parameters of Large Language Models (LLMs) at runtime based on signals such as task descriptions or additional…
Thinking Without Words: Efficient Latent Reasoning with Abstract Chain-of-Thought
Keshav Ramji, Tahira Naseem, Ramón Fernandez Astudillo
While long, explicit chains-of-thought (CoT) have proven effective on complex reasoning tasks, they are costly to generate during inference. Non-verbal reasoning methods have emerg…
DRBENCHER: Can Your Agent Identify the Entity, Retrieve Its Properties and Do the Math?
Young-Suk Lee, Ramon Fernandez Astudillo, Radu Florian
Deep research agents increasingly interleave web browsing with multi-step computation, yet existing benchmarks evaluate these capabilities in isolation, creating a blind spot in as…
Do LLMs Benefit From Their Own Words?
Jenny Y. Huang, Leshem Choshen, Wei Sun +5
In multi-turn conversations, large language models typically condition on the full conversation history: both past user prompts and assistant responses. We revisit this design choi…
Insertion Based Sequence Generation with Learnable Order Dynamics
Dhruvesh Patel, Benjamin Rozonoyer, Gaurav Pandey +3
Existing insertion-based masked diffusion models that generate sequences by interleaving token insertion with unmasking use fixed schedules that are not dependent on the data. For…