3 accepted papers
Large reasoning models achieve strong performance through test-time scaling, but this incurs substantial computational overhead due to long decoding from short prompts. While sparse attention can reduce latency and memory usage, existing methods often degrade reasoning accuracy because selection err…
Large language models (LLMs) have driven significant advancements across diverse NLP tasks, with long-context models gaining prominence for handling extended inputs. However, the expanding key-value (KV) cache size required by Transformer architectures intensifies the memory constraints, particularl…
This paper introduces RaLMSpec, a framework that accelerates iterative retrieval-augmented language model (RaLM) with *speculative retrieval* and *batched verification*. RaLMSpec further introduces several important systems optimizations, including prefetching, optimal speculation stride scheduler,…