RLCD
PulseAugur coverage of RLCD — every cluster mentioning RLCD across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
Jev decision model rivals LLMs in accuracy, speed, and cost
A new decision model called Jev, developed by TypeSafe AI, has been benchmarked against several LLMs for accuracy, speed, and cost in classifying customer support messages. In a test using the Banking77 dataset, Jev 1.1…
-
TypeSafe.ai releases Jev 1.14 with image input and 262k context · 4 sources tracked
TypeSafe.ai has launched Jev 1.14, an update to its System One decision model, now available via the jev-preview alias. This release introduces three experimental features: image input for vision tasks, a 'Box' type for…
-
Jev AI shifts focus from conversation to reliable enterprise automation
Jev, a new AI model from Typesafe, aims to shift AI's focus from conversational abilities to reliable decision-making for enterprise automation. Unlike traditional LLMs optimized for text generation, Jev is trained usin…
-
TypeSafe AI launches Jev 'System One Model' for structured decisions
TypeSafe AI has launched Jev, a new type of "System One Model" designed for fast, structured decisions rather than generalized intelligence. Unlike traditional LLMs, Jev processes state and questions to output typed, st…
-
New RLCD method improves reasoning model calibration and accuracy
Researchers have developed OpenJev-RLCD, a new implementation for reinforcement learning in calibrated decision-making for reasoning models. This approach samples a rationale and then scores the answer distribution usin…
-
New SBERT2S1 method converts sentence encoders to decision models
Researchers have developed SBERT2S1, a method to convert sentence-transformer encoders into "System One" decision models capable of answering schema-constrained questions in a single pass. The study evaluated these mode…
-
New benchmark JevAdvBench tests RLCD model robustness
Researchers have developed JevAdvBench, a novel benchmark designed to evaluate the robustness of models trained with reinforcement learning for calibrated decisions (RLCD). This benchmark addresses the challenge of meas…
-
Open-source Laya model for Jev architecture surpasses benchmarks
A new open-source model named Laya has been developed, designed for the Jev architecture. This model, with 421 million parameters, utilizes a non-autoregressive decision model approach and was trained on a large, human-…
-
Ex-OpenAI researcher launches Jev, an AI for calibrated decisions
Diogo Almeida, a former OpenAI researcher and co-inventor of ChatGPT and RLHF, has launched a new AI model called Jev through his company TypeSafe AI. This model is designed as a "System One Model" optimized for automat…