Claims Browser

Training with debate where two AIs argue can mitigate reward hacking when using LLM judges for RL training

"We show that training with debate, where two AIs argue against each to convince a judge, can mitigate reward hacking, potentially providing a hopeful direction for scaling up accurate training supervision for fuzzy tasks."

Debate training recovers about 45% of the gap between LLM judge training and ground truth training in terms of peak accuracy

"Debate recovers about 45% of the gap between the peak accuracy of training a single policy with an LLM judge and the peak accuracy of training with the ground truth answers."

A graphical notation adapted from Penrose tensor notation can provide a global view of interpretable AI architectures and map directly to PyTorch einsum code

"This graphical notation gives a global view of an architecture and maps one to one onto PyTorch einsum code."

The diagram of Steerling-8B architecture translates into just 33 lines of PyTorch code

"The diagram yields global insights into the architecture (e.g., showing that Steerling is a residual model), a geometric interpretation of each individual operation, and a direct translation into 33 lines of PyTorch code."

HiRA-CAM outperforms both LayerCAM and Grad-CAM on creating useful saliency maps for object classification

"we propose an improved method named HiRA-CAM, and show that it outperforms both LayerCAM and Grad-CAM on creating useful saliency maps for object classification."

Agentic use has risen to 64% of all OpenAI tokens, up from almost none a year ago

"Agentic use now has risen to 64% of all OpenAI tokens, up from almost none a year ago."

Advanced skills like finding software vulnerabilities require carrying long causal chains of 20+ inference steps and do not live in total parameter count once a knowledge-holding threshold is reached

"He notes that advanced skills (e.g., finding software vulnerabilities) are not retrieval/memorization problems. They require carrying long causal chains (20+ inference steps) without losing the thread. This ability does not live in total parameter count once a certain knowledge-holding threshold is reached."

Ornith-1.5 achieves end-to-end self-improvement where the model proposes tasks, generates scaffolds, and produces RL rollouts to create new training experiences

"The headline claim is end-to-end self-improvement: the model proposes tasks, generates scaffolds, and produces RL rollouts to create new training experiences."

A restarted e-detector using a bank of unwindowed betting supermartingales at geometrically spaced restart times requires only O(log t) memory while preserving lifetime anytime validity

"Recency without sacrificing the guarantee is obtained by a restarted e-detector: a bank of unwindowed betting supermartingales at geometrically spaced restart times (O(log t) memory), with the error budget spent over restart instances, which preserves lifetime anytime validity"

On the INSECTS benchmark, the instance-accounted restarted bank achieves zero false spawns and zero false reuses after switches with 0.675 accuracy on INSECTS-reoccurring

"On synthetic multi-concept streams, Electricity, Covertype, and the recurrence-heavy INSECTS benchmark, the instance-accounted restarted bank achieves zero false spawns and zero false reuses after switches and matches or exceeds the retired windowed heuristic (INSECTS-reoccurring accuracy 0.675)"

A spiking neural model combined with Bayesian inference principles creates a brain-like control algorithm capable of operating in uncertain environments

"By combining a biologically inspired spiking neural model with Bayesian inference principles, we propose a brain-like control algorithm capable of operating in uncertain environments."

OpenAI has paused development while implementing new safeguards and diagnosing problems following the HuggingFace attack

"OpenAI has now taken initial steps to address What Happened leading up to HuggingFace attack, including pauses to development while new safeguards are put in place and problems are diagnosed."

Anthropic is preparing for an IPO and their revenue continues to climb, though growth has slowed recently

"Anthropic revenue continues to climb as they prepare for their IPO, although growth has slowed somewhat recently."

A Petri net description of biological neural circuits enables deadline-guaranteed real-time execution with formally verifiable timing semantics

"Here, we propose a Petri net description of biological neural circuits that overcomes these limitations by modeling neurons, synapses, and spike events as a T-timed Petri net with formally verifiable timing semantics, enabling deadline-guaranteed real-time execution"

The Petri neuron reproduces expected dynamical signatures while providing formally bounded timing guarantees with worst-case response times matching analytical predictions across all tested cases

"The Petri neuron reproduces the expected dynamical signatures of each circuit while providing formally bounded timing guarantees throughout, with worst-case response times matching analytical predictions across all three cases."

A local approximation of neural network output with respect to perturbations in inputs and parameters is exact for perturbations that preserve the network activation pattern

"This approximation is exact for perturbations that preserve the network activation pattern, and it allows us to compute analytical expressions for the probability density function and characteristic function of the network output, together with closed-form approximations for its mean and covariance."

Numerical experiments on the Lorenz-63 system and Kuramoto-Sivashinsky equation demonstrate accurate uncertainty propagation through the predictability horizon for high-dimensional dynamical systems

"Numerical experiments on the Lorenz-63 system and the Kuramoto-Sivashinsky equation demonstrate accurate uncertainty propagation through the predictability horizon and the applicability of the proposed framework to high-dimensional dynamical systems."

OpenAI is reportedly building devices for 24x7 monitoring that will upload everything you say to the cloud for AI analysis

"They are reportedly building devices for 24x7 monitoring, no doubt uploading everything you say to the cloud for AI analysis"

The proposed QHAdamW optimizer improves convergence, generalization, and forecasting performance compared to Adam for AQI forecasting using ANNs

"The proposed QHAdamW optimizer addresses the issues on convergence, generalization, and forecasting performance of Adam."