activity
20242026
collaborators
Showing 2025Show all

10 papers · 1 filter

cs.LG2025

MoE-CAP: Benchmarking Cost, Accuracy and Performance of Sparse Mixture-of-Experts Systems

Yinsicheng Jiang, Yao Fu, Yeqi Huang +13

The sparse Mixture-of-Experts (MoE) architecture is increasingly favored for scaling Large Language Models (LLMs) efficiently, but it depends on heterogeneous compute and memory re…

cs.LG2025

Inference-Time Hyper-Scaling with KV Cache Compression

Adrian Łańcucki, Konrad Staniszewski, Piotr Nawrot +1

Inference-time scaling trades efficiency for increased reasoning accuracy by generating longer or more parallel sequences. However, in Transformer LLMs, generation cost is bottlene…

cs.LG2025

Neurosymbolic Diffusion Models

Emile van Krieken, Pasquale Minervini, Edoardo Ponti +1

Neurosymbolic (NeSy) predictors combine neural perception with symbolic reasoning to solve tasks like visual reasoning. However, standard NeSy predictors assume conditional indepen…

cs.CL2025

Zero-Shot Tokenizer Transfer

Benjamin Minixhofer, Edoardo Maria Ponti, Ivan Vulić

Language models (LMs) are bound to their tokenizer, which maps raw text to a sequence of vocabulary items (tokens). This restricts their flexibility: for example, LMs trained prima…

cs.CL2025

Universal Cross-Tokenizer Distillation via Approximate Likelihood Matching

Benjamin Minixhofer, Ivan Vulić, Edoardo Maria Ponti

Distillation has shown remarkable success in transferring knowledge from a Large Language Model (LLM) teacher to a student LLM. However, current distillation methods require simila…

cs.AI2025

Post-hoc Reward Calibration: A Case Study on Length Bias

Zeyu Huang, Zihan Qiu, Zili Wang +2

Reinforcement Learning from Human Feedback aligns the outputs of Large Language Models with human values and preferences. Central to this process is the reward model (RM), which tr…