Recently, Drifting Models and Wasserstein Gradient Flows have attracted substantial attention because they move iterative distributional refinement to training and amortize it into a generator, enabling fast inference. However, existing formulations have been developed largely for continuous Euclidean domains, such as …
Long-form thinking traces can substantially improve the multi-step reasoning performance of large language models (LLMs), but they introduce high inference-time overhead, with latency dominated by sequential decoding. We propose HyperThink, a text-to-parameter approach that amortizes this reasoning computation into a s …
Long-context inference with Large Language Models (LLMs) is bottlenecked by the linearly growing memory of the key-value (KV) cache. Existing compression methods reduce the cache through token eviction or approximation, but degrade sharply at aggressive compression budgets. We propose PatchKV, a training-free framework …