Stores speaker-labeled verbatim messages alongside derived person- and group-level states to improve attribution and long-term memory in multi-party dialogues. Proposes a dual-track memory and a speaker-conditioned Writer trained with SpeakerLevenshtein and GRPO; yields consistent gains on multiple dialogue memory benchmarks.
Uses diffusion-model forking moments as a proxy for perceptual distance to automatically generate pointwise reference-grounded labels, enabling annotation-free training of reference-based image quality assessment metrics.
Shows that Transformer LLMs output a superposition of next-token distributions when token embeddings from distinct text streams are linearly combined, finds this property is architectural but weakened by pretraining, and demonstrates it can be restored by lightweight fine-tuning and disentangled at decode time.
Shows that private post-training changes leave measurable "behavioral shadows" in task-unrelated single-word outputs and proposes Active Taskless Distillation (ATD) to transfer capabilities to a student using only those single-word teacher responses, without teacher logits or parameters.
Trains decoders and diffusion generators to be robust to random subsets of pretrained encoder layers by regularizing layer-fusion during training, narrowing the reconstruction–generation gap in representation autoencoders and improving ImageNet-256 PSNR and gFID.
Reduces self-attention complexity to O(N log N) by using a coarse-to-fine (pyramid) Top-K block selection with LogSumExp scoring, implemented with hardware-aware Triton kernels for fused routing and scoring—aimed at long-context LMs and retrieval tasks.
Evaluates multimodal memory in large audio language models by testing four acoustic evidence types and four memory operations across multi-session spoken histories, stratified across 8K–64K token context budgets.
Introduces an open-source multi-agent harness that automatically constructs, composes, and evolves modular model–harness units for long‑horizon, cross‑domain workflows. Key ingredients include a Host Agent for task decomposition and orchestration, EverOS for durable memory, and a Skill Forge of reusable procedures to improve task coverage via composition.
Learns dynamics-aligned latent representations for parametric PDE forecasting using masked-latent predictive pretraining, a geometry projector that aligns latent trajectory geometry, and a physics-structured latent predictor that separates shared evolution from parameter-dependent responses.
Investigates how rollout policy, token-level KL direction, and learning rate each affect LLM distillation across Llama3 and Qwen2.5 on reasoning tasks; finds KL direction and learning rate dominate outcomes while rollout policy has a modest effect.
Uses 360° panoramic observations to improve vision-and-language navigation by predicting longer action sequences, confidence-guided execution, and combined semantic–geometric panorama features; yields large SR gains on R2R-CE and RxR-CE Val-Unseen.
Records structural priors with skill-specific policies so a runtime agent can select and compose the version of each skill best suited to new states, improving out-of-distribution and compositional generalization for robot manipulation from few demonstrations.