Provides a realtime audio–language foundation model for fully duplex spoken interaction that listens, interprets acoustic cues, reasons, and speaks in parallel. Key features include Deep Perception for acoustic understanding, Seamless Duplex for synchronized streams, and Think‑While‑Speaking for parallel private reasoning; reports strong benchmark results.
Reframes Policy Mirror Descent for autoregressive LLM generation into a critic-free, trajectory-level objective using the Bellman equations to avoid per-token value estimation. Derives a practical token-level loss with a smoothed mismatch-correction weight and shows improved math-reasoning accuracy on benchmarks.
AI Agent Papers2026
Estimates a model's probability of correctness by leveraging its past graded episodes: XConf recalls similar past cases and prompts the model to reflect and restate confidence, improving calibration and discrimination across reasoning, coding, multimodal QA, and agent tasks at much lower generation cost.
AI Model2026
Compresses a 27B-class multimodal model into end-to-end ternary weights to run 27B reasoning on-device: 5.9–8.6 GB deployed footprint, 262K-token context, ~98.2% of FP16 benchmark performance; ships MLX and GGUF packs and runs on Apple MLX and CUDA.
AI Video Papers2026
Haoyu Zhao, Zihao Zhao +11Affiliation: National University of Singapore, Affiliation: Fudan University +1
Evaluates whether MiniMax-H3 can infer latent physical events from incomplete multimodal evidence and produce videos consistent with those inferences. Builds a 517-instance benchmark across four reasoning scenarios (multi-view, audio disambiguation, video continuation, audiovisual integration) and quantifies modality-specific failure modes.
Analyzes a failure mode in PPO critics called “Value Flattening” and proposes SP³O (Sparse PPO), which supervises a small set of well-separated states per response to mitigate flattened value estimates and improve policy learning for LLM alignment and reasoning tasks.
Analyzes why on-policy distillation can produce excessively long student rollouts by identifying termination-token mismatch between students and teachers, and proposes aggregating functionally equivalent EOS tokens as a shared semantic stopping action to mitigate length inflation.
Teaches vision-language models to predict and integrate physical-world state transitions from interaction trajectories (observation → action → next observation). Introduces a three-level curriculum and the LSI-108K dataset, and applies supervised fine-tuning plus on-policy distillation to improve local transition modeling and long-horizon spatial integration.
AI Dataset2026
A multimodal supervised fine‑tuning corpus for training MLLMs: 44.3M samples (100.3M QA pairs) over 19.7M image groups, annotated with short/long chain‑of‑thought and produced by rerunning the HoneyPipe curation pipeline with multi‑model rollouts and verifications.
AI Dataset2026
Provides 2,000 synthetic multiple-choice items designed for continuation log-likelihood scoring to evaluate small language models' Theory of Mind (social-cognitive) abilities; 40 constructs, balanced answer positions, and easy/medium difficulty.
AI Model2026
Provides compact mixed-precision GGUF quantizations of UkisAI's Swift 1.5 (derived from Qwen3.8-27B), using ISTA GSQ-RCO per-tensor allocations with Swift-specific refinement. Offers multiple 8–12 GB tiers, optional MTP heads, and KLD evaluation against the Swift BF16 baseline.
AI Dataset2026
Provides 150,000 source‑grounded decision examples for training models that pick options, judge yes/no propositions, or assign ordered scores. Each row pairs a 'state' with JEV-style typed questions (CHOICE/NOUL/SCORE); multiple configs and train/test splits are included, license mixed/unknown.