Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling
arxiv.orgยท21h
๐Ÿ”Query Compilers
How to Teach Large Multimodal Models New Skills
arxiv.orgยท21h
๐Ÿ—๏ธAI Infrastructure
Own your AI: Learn how to fine-tune Gemma 3 270M and run it on-device
developers.googleblog.comยท2dยท
Discuss: Hacker News
๐Ÿ—ฃ๏ธSpeech Synthesis
Memory Retrieval and Consolidation in Large Language Models through Function Tokens
arxiv.orgยท21h
ฮปFunctional Programming
Self-Improving LLM Agents at Test-Time
arxiv.orgยท21h
๐Ÿค–AI agents
Distilling Reasoning into Student LLMs: Local Naturalness for Selecting Teacher Data
arxiv.orgยท3d
๐ŸคFederated Learning
Exponential Error Bounds for Information Bottleneck Source Coding Problems
arxiv.orgยท21h
โฑ๏ธTime-series Optimization
MARC: Memory-Augmented RL Token Compression for Efficient Video Understanding
arxiv.orgยท21h
๐Ÿ—๏ธAI Infrastructure
Why LLMs cannot reach GenAI, but why it looked like they could
haversine.substack.comยท2hยท
Discuss: Substack
๐Ÿ—๏ธAI Infrastructure
Activation Alchemist: Sculpting Stability with Functional Signatures
dev.toยท5hยท
Discuss: DEV
ฮปFunctional Programming
Learning from Failures: Understanding LLM Alignment through Failure-Aware Inverse RL
arxiv.orgยท2d
๐Ÿ Self-hosted AI
94% of Developers Waste Tokens on Reasoning LLMs. Here's Why.
dev.toยท19hยท
Discuss: DEV
๐Ÿค–AI agents
Optimal Stopping in Latent Diffusion Models
arxiv.orgยท21h
๐Ÿ“ฑEdge AI
Aligning Large Language Models via Fully Self-Synthetic Data
arxiv.orgยท1d
๐Ÿ Self-hosted AI
Distribution Preference Optimization: A Fine-grained Perspective for LLM Unlearning
arxiv.orgยท3d
๐Ÿ Self-hosted AI
Revisiting Mixout: An Overlooked Path to Robust Finetuning
arxiv.orgยท1d
๐Ÿ“ฑEdge AI
Decentralized Intelligence: Empowering Autonomous Systems with Localized Learning by Arvind Sundararajan
dev.toยท1dยท
Discuss: DEV
๐ŸคFederated Learning
Get RICH or Die Scaling: Profitably Trading Inference Compute for Robustness
arxiv.orgยท1d
๐Ÿ“ฑEdge AI
ECLipsE-Gen-Local: Efficient Compositional Local Lipschitz Estimates for Deep Neural Networks
arxiv.orgยท2d
๐Ÿ“ฑEdge AI
Can Speech LLMs Think while Listening?
arxiv.orgยท21h
๐ŸŽ™๏ธWhisper