We have 6 of 31 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
ESSA (Evolutionary Strategies for Scalable Alignment), a gradient-free online alignment stage that follows supervised fine-tuning (SFT) and replaces the gradient loop with inference-only black-box search, is presented.
A difficulty-aware scaling coefficient is proposed, inspired by Focal loss, that down-weights updates on high-success sampled groups and improves average math pass@256 in all three cases, without increasing group size or computational cost.
KronSAE is proposed, a design that factorizes the latent space into heads and forms post-latent features as pairwise compositions of lower-dimensional pre-latents using mAND, a differentiable AND-like interaction that imposes a compositional co-activation prior while remaining compatible with standard SAE objectives an...
Layer-Integrated Memory (LIMe) is introduced, a lightweight extension that leverages existing key-value buffers and learns per-head, per-layer routing weights to integrate representations from previous layers to improve perplexity per FLOP and yield strong gains on synthetic tasks while preserving higher value-vector e...
This work introduces Soft Latent Thinking, a method that replaces the LM head during reasoning with a lightweight projector, enabling autoregressive rollout in embedding space where reasoning steps remain continuous rather than tokenized.
Qantara is presented, an end-to-end JEPA whose joint training objective pairs a Brownian-bridge interpolant between consecutive clean latents on the state axis with noise-to-data flow matching on the action axis, and these results move JEPA world models from single-paradigm planners to multi-paradigm controllers.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
