Search and filter through extracted claims from AI researchers.
Showing 1661-1680 of 2685 claims of type "fact"
The assumption that better corpus models improve predictions of human processing difficulty has been undermined by recent empirical work showing better corpus models can be worse predictors.
TimePNS improves time-series explanations by incorporating necessity signals through counterfactual interventions, not just sufficiency.
Task-specific fine-tuning substantially improves open-source LLMs on medical narrative generation tasks and narrows the gap with commercial models.
MedGame transforms static clinical cases into structured, executable storytelling games using a dual-engine design with Medical Narrative Designer and Story Director.
Metric saliency propagates through the image formation process itself, including multi-bounce light transport, capturing parameter dependencies that are semi-opaque to manual inspection.
Differentiable renderers can produce metric saliency maps that identify which scene elements most influence a scalar metric, analogous to gradient-based saliency in neural networks.
Pretraining EnsembleEGNN on conformational ensembles dramatically improves molecular property prediction compared to training from scratch (R²=0.477 vs 0.005).
LLM judgment revision is structured along three dimensions paralleling human social psychology: distance between incoming view and initial position, source attribution, and coalition structure.
OpenForgeRL enables training harness-based agents end-to-end at scale by decoupling training and inference through a lightweight proxy and Kubernetes orchestrator.
Complex inference harnesses like Claude Code, Codex, and OpenClaw make AI agents hard to train end-to-end because standard SFT/RL stacks cannot natively express stateful, multi-process harness inference.
Visual Contrastive Self-Distillation (VCSD) can perform on-policy self-distillation without requiring privileged answers or visual evidence, using only input conditioning through image-content removal.
Block Attention Residuals can boost deep-layer effective rank by approximately 12% by routing completed block summaries into later linear layers
Hybrid Linear-Softmax Attention combining gated linear attention for O(N)-dominated mixing with periodic gated-softmax anchors can avoid quadratic attention while restoring full-rank token interactions that pure linear attention lacks
SANA-Video 2.0 can generate high-quality video up to 720p on a single GPU while matching full-softmax video DiTs in quality and retaining favorable long-sequence scaling of linear attention
Cross-modal on-policy distillation can transfer reasoning capabilities from text teachers to audio-language students, extending beyond textually recoverable content to reasoning grounded in non-linguistic events, prosody, and conversation
Large audio-language models still lag behind text-based large language models in deep logical reasoning, primarily due to the scarcity of high-quality audio reasoning data
Multimodal approaches often outperform single modality approaches in downstream tasks because different modalities provide complementary information
Uncertainty-weighted optimal transport can ensure robust knowledge transfer by dynamically weighting feature-level alignment based on prediction confidence to suppress noisy supervision
Diffusion-based reconstruction combined with occlusion-type identification can address the limitation of degraded iris recognition when substantial texture is corrupted
Iris recognition performance degrades when discriminative iris texture is partially occluded by eyelids, eyelashes, specular reflections, or other acquisition artifacts
Pipeline data may be stale or degraded.
Last synthesis: 2026-09-20. 8,975 pending.