4 papers
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…
Latency and Token-Aware Test-Time Compute
Jenny Y. Huang, Mehul Damani, Yousef El-Kurdi +2
Inference-time scaling has emerged as a powerful way to improve large language model (LLM) performance by generating multiple candidate responses and selecting among them. However,…
Optimal Policy Minimum Bayesian Risk
Ramón Fernandez Astudillo, Md Arafat Sultan, Aashka Trivedi +4
Inference scaling helps LLMs solve complex reasoning problems through extended runtime computation. On top of long chain-of-thought (long-CoT) models, purely inference-time techniq…
Ensemble-Instruct: Generating Instruction-Tuning Data with a Heterogeneous Mixture of LMs
Young-Suk Lee, Md Arafat Sultan, Yousef El-Kurdi +4
Using in-context learning (ICL) for data generation, techniques such as Self-Instruct (Wang et al., 2023) or the follow-up Alpaca (Taori et al., 2023) can train strong conversation…