The Archive

Jane Street has installed Etched's first shipped AI cluster system, and was so impressed, it led another massive round, the startup says.

Analysis of why GPT-style tokenization fails for symbolic music: compression requires domain-specific coordinate systems.

Reproduction and sensitivity analysis of WEASEL 2.0 time series classifier with empirical evaluation of hyperparameter rules.

LLM-based framework for explaining flight safety events from pilot behavior, addressing modal inconsistency challenges.

IOL-AI Challenge: open competition on linguistic reasoning (unseen IOL 2026 problems) with expert human evaluation.

Theoretical framework for LLM reward shaping in hybrid RL agents using bounded potential functions preserving optimal policy.

Flow-matching energy-based models enable physics-informed generative modeling of PDE fields with composable energies for OOD detection and inversion.

Uniform Manifold Approximation and Projection (UMAP) is a dimensionality reduction technique widely used for visualization and feature extraction. Applications... Uniform Manifold Approximation and Projection (UMAP) is a dimensionality reduction technique widely used for visualization and feature extraction. Applications range across exploratory data analysis, topic modeling, and single-cell analysis. Many of these workflows are iterative and exploratory, requiring UMAP to be run repeatedly as users analyze their data or tune parameters. As datasets grow… Source

Traceable Trust framework proposes structured assessment process for AI-driven laboratory decision-making in bioscience to ensure reproducibility and accountability.

LLM-as-judge framework with abstention mechanism and provable risk bounds improves reliability on factual tasks via uncertainty quantification and retrieval.

TSN4PI framework tracks political ideology evolution on social media using LLMs with style transfer and unsupervised domain adaptation to address annotation scarcity.

Source-free domain adaptation for ultrasound tongue segmentation under extreme data scarcity using pseudo-label refinement and reliability filtering.

Recirculation inference-time technique adds learned recurrence to transformers enabling stateful belief tracking, reducing perplexity with minimal latency overhead.

AVShift benchmark evaluates authorship verification robustness across genre, temporal, and AI-assisted writing distribution shifts in German text (150K+ examples).

Comparative study of ML and deep learning crop-yield forecasting methods under extreme drought conditions using 16 meteorological drivers.

LM-based log anomaly detectors show miscalibrated confidence on anomalous events; calibration techniques fail to capture prediction reliability under class imbalance.

Apple’s leaked camera-equipped AirPods might avoid the privacy pitfalls of other AI wearables by preventing users from recording photos and videos.

Neurosymbolic world model decouples reward prediction from structured symbolic latent state components, enabling zero-shot transfer across RL tasks.

Self-supervised pre-training on single tabular datasets yields strong transfer learning for foundation models without massive synthetic corpora.

Code world models risk exploitation in continuous control when sampling-based acceptance fails to detect rare critical events.

LLM hidden state manifolds organize as small-world networks enabling multi-hop reasoning; bypasses attention artifacts via geometric analysis.

SIGMA uses SHAP guidance for LLM-based automated feature engineering without metadata, reducing context window overflow and local optima.

Continual Abstraction Discovery evolves procedural content generators via LLM mutation, extracting reusable primitives across game level design.

Graph-structured online difficulty estimation optimizes RLVR exploration budgets for LLM reasoning by adaptive sample-difficulty scheduling.

Small models grade open-ended exam answers reliably as frontier models when given explicit rubrics, reducing inference cost 80%+.

EvoTS-Agent self-evolves via validation feedback for autonomous financial time-series change-point detection without manual model selection.

Collective Counterfactual Planning formalizes multi-agent coordination where representational geometry constrains perception and goal verification.

SpeechSense dataset captures paralinguistic cues for speech sentiment analysis, preserving acoustic features beyond ASR transcription.

Adaptive policy portfolios optimize robustness in MDPs by synthesizing offline policies with lightweight online selection, reducing conservative bounds when environment dynamics become identifiable.

Decentralized planning framework for lifelong multi-agent path finding that improves computational scalability over RHCR through group-based MDP theory.