“Paper Bitcoin” falls, is it time for BTC to rise higher?
Even though bulls face headwinds, Willy Woo, an on-chain analyst, is bullish on bitcoin. He cites recent developments around spot, ...
Tag: paper
This Microsoft machine learning paper proposes ChunkAttention: a novel self-attention module to efficiently manage the KV cache and accelerate the self-attention kernel for LLM inference
The development of large language models (LLM) in artificial intelligence represents an important advance. These models underpin many of today's ...
This AI paper proposes a basic interactive agent model that uses a new multitasking agent training paradigm to train AI agents on a wide range of domains, data sets, and tasks.
ai development is moving from static task-focused models to dynamic and adaptive agent-based systems suitable for various applications. ai systems ...
This AI paper proposes Infini-Gram: an innovative approach to scaling and improving N-Gram models beyond traditional limits
Pretrained on trillion-token corpuses, large neural language models (LLMs) have made notable gains in performance (Touvron et al., 2023a; Geng ...
This China AI paper proposes a small and efficient model for optical flow estimation
Optical flow estimation, a cornerstone of computer vision, allows predicting motion on a per-pixel basis between consecutive images. This technology ...
This AI paper from Stanford and Google DeepMind reveals how efficient exploration increases the effectiveness of human feedback to improve large language models
artificial intelligence has seen notable advances with the development of large language models (LLM). Thanks to techniques such as reinforcement ...
This AI paper from UT Austin and JPMorgan Chase presents a novel algorithm for machine unlearning in generative image-to-image models
In an era where digital privacy has become paramount, the ability of artificial intelligence (ai) systems to forget specific data ...
This Apple AI paper proposes fusing acoustic models to dramatically reduce word error rates in speech recognition systems
Significant improvements have been made to improve the accuracy and efficiency of automatic speech recognition (ASR) systems. Recent research delves ...
This China AI paper introduces 'AGENTBOARD': an open source evaluation framework tailored to the analytical evaluation of multi-shift LLM agents
Evaluating LLMs as versatile agents is crucial for their integration into practical applications. However, existing evaluation frameworks face challenges in ...
This AI paper from ETH Zurich, Google and Max Plank proposes an effective AI strategy to boost the performance of reward models for RLHF (reinforcement learning from human feedback)
In language model alignment, the effectiveness of reinforcement learning from human feedback (RLHF) depends on the excellence of the underlying ...
Recommended.
