| tech.blog / monetary.live / ivan.digital
September 08, 2026 at 18:14 UTC
🤗 HuggingFace Daily Papers
Dr. Claw: An AI Scientist Workspace for Vibe Research
▲ 123 🎓 Lehigh University
  • Command-line coding agents (e.g., Claude Code, Gemini CLI) can already read and write files and sustain long sessions, yet end-to-end research still f...
  • Claw, an open-source workspace that wraps existing coding-agent executors in a controllable and auditable human-in-the-loop workflow rather than intro...
  • Persistent state objects, a reusable skill library, and multi-executor coordination link human decisions to AI execution, turning planning, execution,...
  • Claw through an interactive three-view scenario and a failure-recovery walkthrough, and evaluate it against a bare command-line agent sharing the same...
Unlocking Lossless Speedups in LLMs via Discrete Diffusion
▲ 64 🎓 Institute of Foundation Models
  • Diffusion-augmented autoregressive language models use parallel token sampling via distilled diffusion weights and a specialized sampler to accelerate...
FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience
▲ 48 🏢 Tencent Hunyuan
  • FlowBalance improves reasoning models via verifier-calibrated self-guidance using trajectory-level score reweighting and profile-based trajectory bala...
Motion-Omni: End-to-End Joint Speech and Full-Body Motion for Spoken Dialogue
▲ 38 🎓 Peking University
  • Motion-Omni is an end-to-end framework that jointly generates spoken dialogue and full-body co-speech motion from shared hidden states, using scalable...
The Attention Triangle in Audio-Video Models
▲ 28 🎓 Tel Aviv University
  • Audio-video diffusion models exhibit bidirectional semantic leakage through cross-modal attention pathways, which can be diagnosed via attention-deriv...
Enoki: Efficient Multi-Level Hallucination Detection
▲ 19 🎓 s-nlp
  • Enoki is an open information extraction framework that unifies claim-level verification and span-level hallucination localization through shared relat...
ENEAS: Embedding-guided Neural Ensemble for Adaptive Segmentation
▲ 18 🏢 SperidLabs
  • ENEAS unifies text-prompted instance tracking and open-concept semantic discovery via temporal memory extension and a verification layer that filters ...
Bilevel Coordination Game 1Approximate Potential Game 1Stochastic Movement 1Semantic Memory States 1Falsifiable Persistent-Harm Condition 1Information-Theoretic Impossibility 1Text-Indistinguishable Environments 1Environment-Grounded Gate 1Stochastic Reflective Memory Ascent 1Evaluation Risk 1Non-Degenerate Corrective Mass 1Piecewise-Stationary Environments 1
🏛️ Top Research Institutions
Alignment-Free Text-Audiobox for Voice Dubbing and Full-Duplex Dialogue Synthesis
🏛️ FAIR at Meta Other CS
  • High-quality voice dubbing and full-duplex dialogue synthesis require complex alignment procedures that limit scalabilit...
  • Alignment-Free Text-Audiobox framework using Diffusion Transformer with flow-matching training for unified voice dubbing...
Global Multi-Maturity SPX-VIX Calibration Beyond Markovian Stitching
🏛️ JPMorganChase Quantitative Finance
  • Joint calibration of SPX and VIX smiles across maturities is constrained by unrealistic Markovian independence assumptio...
  • Global multi-maturity calibration framework that removes conditional-independence restrictions while maintaining exact l...
Does p-Hacking Mitigate or Exacerbate the Effects of Publication Bias?
🏛️ University of Wisconsin-Madison Economics
  • ** Randomization-based inference for treatment effects in matched-pair studies with binary outcomes lacks exact, distrib...
  • ** Develops exact randomization inference method requiring neither monotonicity assumptions nor distributional specifica...
AI & Machine Learning
396 papers 7 cats
Systems & Infrastructure
26 papers 6 cats
Software & Programming
23 papers 4 cats
Theory & Algorithms
32 papers 5 cats
Applications
99 papers 7 cats
Other CS
30 papers 9 cats
Quantitative Finance
9 papers 9 cats
Economics
5 papers 3 cats