Latest in AI and developer tools

Today's trail

updated 14m ago

Last hour Today

Pulse · msn.com

msn.com · Nov 3

Taiwan's AI stock boom is making early-stage startups a harder sell

Taiwan's AI stock boom could be making it harder for young startups to win their first funding checks.

Pulse · jetstream.blog

jetstream.blog · Nov 3

Data#3 AI Practice: $3.4B Sales, FY27 Pods [2026]

Google has released the October 2026 edition of "Google AI Updates" (The latest AI news we announced in October 2026), summarizing the latest AI-related news announced last month in October 2026. The highlights for October 2026 include the general availability of the cutting-edge AI model "Gemini 4 Argon," the announcement of "Gemini 4 Carbon" (Google's most powerful AI model to date) just two weeks later, and "Genie 4," a real-time interactive world model with native audio support. Other…

Pulse

Top picks this weekarXiv

arXiv

Nov 2

Whitened Advantage Summation Policy Optimization: Correlation-Aware Aggregation for Multi-Reward Reinforcement Learning

Reinforcement learning for language models often combines several rewards by normalizing each one and adding them with fixed weights. When the rewards are correlated, the variation they share is counted once for every reward that carries it. In terms of the ZCA-decorrelated components of the reward signal, the weighted sum applies weights $Σ^{1/2} w$ rather than the chosen weights $w$, where $Σ$ is the reward correlation matrix. We study WAS-PO, which whitens the normalized advantages with $Σ^{-1/2}$ before adding them, so that the chosen weights apply to those components. We derive the scalarized objective whose gradient it follows when normalization statistics are held fixed, show that its weights on the raw rewards can be negative, give an exact two-reward condition under which the lighter-weighted reward is reversed, and give a loose bound on the error from estimating $Σ$ during training. On tool calling with Qwen2.5-1.5B-Instruct, WAS-PO reached a higher mean training reward than GDPO in all seven seeds with three rewards (3.8% higher) and all three seeds with four (3.9% higher). WAS-PO also replaces GDPO's per-prompt normalization with batch-level normalization, and we did…

Top picks

All arXiv Hugging Face GitHub Hacker News

Latest Top this week

arXiv

Sep 21

An Unexpected Robot Policy: Early Evaluations of GPT-6 Astra on RoboDojo and Beyond

Embodied AI systems are often organized into System 1 and System 2. System 1 is typically a pretrained policy that generates actions at high frequency, whereas System 2 is often instantiated as a vision-enabled language model for high-level planning. We ask whether a large…

arXiv

Sep 21

APEXA: Execution-Integrity Enforcement for Multi-Agent LLM Automation of Synchrotron Data Reduction

LLM agents can make expert scientific workflows accessible through natural language, but a plausible response does not establish that the requested computation ran. This matters at synchrotron facilities, where calibration and integration are multi-step operations over…

arXiv

Sep 21

MCP-GRANITE Benchmark: GRANularity Interface TEsting for MCP-Based LLM Agents

As LLM agents increasingly interact with external tools through standardized protocols such as MCP, tool-interface design becomes a critical yet underexplored factor. How funψtionality is decomposed into tools affects whether an agent can select the right tool and construct…

arXiv

Sep 21

Relightable 3D Avatar Reconstruction with Semantic-Adaptive Motion-Illumination Responses

Reconstructing expressive and relightable 3D head avatars from monocular videos remains challenging in computer vision, as it requires accurate modeling of both non-rigid facial motion and illumination-dependent appearance. Existing Gaussian avatar methods commonly rely on…

This week in Multimodal

This week

Get the morning digest

Today's trail

Pulse

Top picks

Today's trail

Rising tools

This week in Multimodal

Most active this week