A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Fei Wang, Xingchen Wan, Ruoxi Sun, et al.
Astute RAG adaptively elicits internal knowledge from LLMs and iteratively consolidates it with external sources to overcome imperfect retrieval in RAG.
Zhisheng Hu, Minghui Zhu, Peng Liu
Proposes an adaptive cyber defense using learning-based POMDP with Thompson sampling to counter multi-stage attacks under unknown exploit likelihoods and impacts.
Jialong Li, Mingyue Zhang, Nianyu Li, et al.
This paper surveys the potential benefits and challenges of using generative AI, particularly LLMs, to enhance self-adaptive systems, and outlines a research roadmap.
Gianni A. Di, Marco Dorigo
AntNet introduces a distributed, stigmergy-based ant colony optimization approach for adaptive routing in communications networks, outperforming six state-of-the-art algorithms.
Diederik P. Kingma, Jimmy Ba
Adam is a computationally efficient stochastic optimization method using adaptive moment estimates, invariant to gradient scaling and suitable for large-scale problems.
Unknown
An adaptive reasoning model that selects reasoning formats based on task difficulty to balance performance and efficiency, trained with Ada-GRPO to prevent format collapse.
Unknown
FastBERT introduces a speed-tunable encoder with adaptive inference time by adding early-exit branches at each transformer layer.
Unknown
This paper proposes adaptive KV cache compression policies for LLMs to reduce memory usage while maintaining generation quality.
Unknown
Gora introduces gradient-driven adaptive low-rank adaptation that matches full fine-tuning performance.
Unknown
Instructflow uses adaptive symbolic constraints to guide code generation for long-horizon robotic planning.
Yuyao Wang, Zhongjian Zhang, Mo Chi, et al.
EvoMemBench benchmarks agent memory from a self-evolving perspective, providing a unified testbed for adaptive memory in execution-oriented tasks.
Sudhir Kumar, Glen Stecher, Michael Suleski, et al.
MEGA12 introduces heuristics to reduce computational time for substitution model selection and bootstrap tests, plus evolutionary sparse learning for fragile clade identification.