This work proposes ParetoTransport, a training-free guidance method for pre-trained flow-matching models that explicitly specifies and refines a population-level distribution in objective space, and establishes a convergence result and demonstrates state-of-the-art performance on standard offline MOO benchmarks.
Stephan Holly, Sepp Hochreiter, Werner Zellinger· 0 citations
KVpop is introduced, which learns a fixed-budget KV eviction policy by directly supervising the keep-or-drop decision, and introduces a delayed memory-based scorer that, uniquely among learned eviction methods, defers scoring for a fixed number of steps to exploit near-future context.
Lukas Hauzenberger, N. Schmidinger, A. Hartl et al.· arXiv.org· 1 citation
This work reformulates the output projection followed by top-k token selection as a maximum inner product search over token embeddings and replaces the dense vocabulary projection with an HNSW-based vector index, suggesting approximate retrieval is a practical alternative to dense output projections in latency-sensitiv...
M. Loretz, Sepp Hochreiter· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.