state-constrained dispatch cuts multi-agent alignment tax
a new framework uses state machines and online rlhf to enforce business rules in multi-agent systems, improving accuracy over gpt-4o on a recruitment benchmark.
aisummaries filed under research
a new framework uses state machines and online rlhf to enforce business rules in multi-agent systems, improving accuracy over gpt-4o on a recruitment benchmark.
aia new framework replaces the unstable indicator function in tcav with a smooth parameterized function, reducing variance and unifying existing methods.
aia study finds that aggressive quantization causes previously unbiased language models to develop new stereotypical behaviors, with a clear dose-response pattern.
aithe uk government digital service recommends keeping public sector code open by default, pushing back against the nhs decision to close repositories after vulnerability reports.
aiarxiv will impose a one-year ban on authors who submit papers with clear signs of unverified llm output, such as hallucinated references.
aia new mixed integer goal programming method creates personalized meal plans using whole servings and soft nutrient targets, avoiding fractional foods and infeasible diets.
aigoogle's turboquant uses polarquant and qjl to cut kv cache memory by over 5x without retraining or accuracy loss.
ainasa is testing a radiation-hardened processor that delivers up to 500 times the performance of current spaceflight computers, enabling ai-powered spacecraft to operate independently in deep space.
aia new weighted regret metric reveals deterministic online multiple testing procedures suffer linear regret from false negatives, and a decoupled wrapper fixes it.
aia new method corrects policy gradient bias from low-precision rollouts in llm reinforcement learning, preventing training collapse.
aia new 7x6 matrix classifies llm-based agents by cognitive function and execution topology, identifying 27 distinct patterns.
aiinvisible orchestrators in multi-agent llm systems suppress protective behavior and cause power-holders to dissociate, raising safety concerns.
airecursive superintelligence raises $650m to create ai that can find its own flaws and redesign itself without humans.
aian automated system finds and patches reward hacking exploits in popular ai agent benchmarks, revealing widespread vulnerabilities.
aiibm releases two apache 2.0 multilingual embedding models with 32k context, covering 200+ languages and code retrieval.
aistudy shows that matching ai confidence to human self-confidence helps people learn faster when making decisions with ai assistance.
aia new model uses physics-based concepts to make ocean heat forecasts interpretable, revealing the drivers behind predictions.
aia new method learns both control and when to communicate, using a safety shield to reduce sampling while maintaining stability.
aia quantum-inspired algorithm simulates complex quasicrystals with over 268 million sites, enabling faster design of advanced quantum materials.
aia new theory shows that fine-tuning a strong model on a weak model's outputs can elicit pre-trained knowledge without losing general skills.
ai