6 things to fix before RLHF turns your biases into features
Your reward model is learning exactly what your annotators prefer. The problem is that "better" and "unbiased" are two different things, and RLHF has no way to tell them apart.
Alibaba’s Qwen Team Launches Qwen3.7-Plus, Adding Vision, Deep Reasoning, Tool Invocation, and Autonomous Iteration on the Bailian Platform
Qwen3.7-Plus is Alibaba's multimodal agent model on Bailian, understanding images and video while adding self-programming and tool invocation.
The post Alibaba’s Qwen Team Launches Qwen3.7-Plus, Adding Vision, Deep Reasoning, Tool Invocation, and Autonomous Iteration on the Bailian Platform appeared...
This article is from Making AI Work, MIT Technology Review’s limited-run newsletter examining how to apply LLMs across industries. To receive it in your inbox,sign up here. From accounting to design to market research and product development, there’s a staggering breadth of skills needed to run a bu...
JetBrains Releases Mellum2: A 12B MoE Model for Fast, Specialized Tasks in Multi-Model AI Pipelines
JetBrains releases Mellum2 under Apache 2.0 — a 12B MoE model trained on 10.6 trillion tokens for AI workflows.
The post JetBrains Releases Mellum2: A 12B MoE Model for Fast, Specialized Tasks in Multi-Model AI Pipelines appeared first on MarkTechPost.
DAStatFormer: A Hybrid Multibranch Transformer with Statistical Feature Integration for DAS-Based Pattern Recognitions
arXiv:2606.00081v1 Announce Type: new
Abstract: Distributed Acoustic Sensing (DAS) enables large-scale monitoring through optical fibers, but its high dimensionality and complex spatio-temporal patterns make event classification demanding. Existing deep learning approaches-CNNs, recurrent models, a...
Hoeffding Concept Bottleneck Models with Applications to Overhead Images
arXiv:2606.00082v1 Announce Type: new
Abstract: Explainability of deep learning algorithms is critical for computer-vision applications with high-stake decisions. Concept bottleneck models (CBM) have recently shown promising performance to provide explainable and accurate predictions for classifica...
From Demonstrations to Rewards: Test-Time Prompt Optimization for VLM Reward Models
arXiv:2606.00083v1 Announce Type: new
Abstract: Reinforcement learning relies on accurate reward functions, which are often hand-crafted or even unavailable in real-world applications, such as robotics. Recent work has explored the zero-shot reasoning capabilities of pre-trained Vision-Language Mod...
A Shared Valence Axis Across Modern LLMs and Human EEG: The Saturation Regularity
arXiv:2606.00129v1 Announce Type: new
Abstract: Large language models (LLMs) have emerged as powerful representation learners whose internal features increasingly align with human cognition. We study whether modern LLMs can serve as a lens for understanding neural representations in the human brain...
Emergent Collaborative Deliberation in Multi-Model AI Systems: A BFT-Derived Protocol for Epistemic Synthesis
arXiv:2606.00005v1 Announce Type: new
Abstract: We present the Consilium Protocol, a Byzantine Fault Tolerance-derived architecture for structured multi-model AI deliberation that treats inter-model disagreement as epistemic signal rather than error. The protocol assigns engineered cognitive person...
Deliberative Curation: A Protocol for Multi-Agent Knowledge Bases
arXiv:2606.00007v1 Announce Type: new
Abstract: As AI agents transition from isolated tools to collaborative participants in shared knowledge ecosystems, governing collective knowledge curation becomes a critical challenge. Human platform governance mechanisms do not transfer directly: agent statel...
Agents on a Tree: Pathwise Coordination for Multi-Objective Molecular Optimization
arXiv:2606.00008v1 Announce Type: new
Abstract: Multi-objective molecular optimization requires searching vast chemical spaces under conflicting objectives, where early design decisions strongly constrain downstream outcomes. Existing methods typically rely on a single policy or fixed scalarization...
NVIDIA Jetson Brings Agentic AI to the Physical World
Agentic AI is getting physical. At COMPUTEX on Tuesday, NVIDIA announced NVIDIA JetPack 7.2 and NVIDIA NemoClaw support on NVIDIA Jetson. JetPack 7.2 brings agentic AI skills, Yocto project support, NVIDIA CUDA 13 on NVIDIA Jetson Orin, a substantial performance gain on Jetson AGX Orin 32GB module a...
Codex is becoming a productivity tool for everyone
The Next Era of Knowledge Work report explores how Codex is transforming productivity through AI-powered research, data analysis, workflow automation, and content creation.
MiniMax Releases MiniMax M3 with MSA Architecture Supporting 1M-Token Context, Native Multimodality, and Agentic Coding
MiniMax M3 introduces MiniMax Sparse Attention, a 1M-token context window, and native image, video, and computer use support.
The post MiniMax Releases MiniMax M3 with MSA Architecture Supporting 1M-Token Context, Native Multimodality, and Agentic Coding appeared first on MarkTechPost.
RAG Is Not Machine Learning, and the ML Toolkit Solves the Wrong Problem
Enterprise Document Intelligence [Vol.1 #3] - Why the ML toolkit (hyperparameter sweeps, train/test splits, explainability frameworks) solves the wrong problem, and what to use instead
The post RAG Is Not Machine Learning, and the ML Toolkit Solves the Wrong Problem appeared first on Towards Data Sc...
Meet Memory OS: A 6-Layer Open-Source Memory Stack Built on Top of Hermes Agent
The open-source project adds local persistent memory to Hermes Agent through six layers, gated retrieval, and a wiki.
The post Meet Memory OS: A 6-Layer Open-Source Memory Stack Built on Top of Hermes Agent appeared first on MarkTechPost.
A collage of I/O-related images, including the Antigravity Coffee Co. pop-up, a colorful jellyfish and a still from the Timmy TPU video. The word AI repeats three times on the left of the image, and there are colorful icons, including a sparkle, as well.