Difference between revisions of "AI research trends"

From GISAXS
Jump to: navigation, search
(→‎Memes)
(→‎Sampling)
 
(4 intermediate revisions by the same user not shown)
Line 52: Line 52:
 
==Long-Term Memory==
 
==Long-Term Memory==
 
* 2025-04: [https://arxiv.org/abs/2504.19413 Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory]
 
* 2025-04: [https://arxiv.org/abs/2504.19413 Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory]
 +
 +
===Storage and Retrieval===
 +
* 2025-09: [https://arxiv.org/abs/2509.04439 ArcMemo: Abstract Reasoning Composition with Lifelong LLM Memory]
  
 
===Episodic Memory===
 
===Episodic Memory===
 
* 2024-03: [https://arxiv.org/abs/2403.11901 Larimar: Large Language Models with Episodic Memory Control]
 
* 2024-03: [https://arxiv.org/abs/2403.11901 Larimar: Large Language Models with Episodic Memory Control]
 +
* 2025-08: [https://arxiv.org/abs/2508.16153 AgentFly: Fine-tuning LLM Agents without Fine-tuning LLMs]
  
 
=Updating Weights at Inference-time=
 
=Updating Weights at Inference-time=
 
* 2025-01: [https://arxiv.org/abs/2501.06252 Transformer<sup>2</sup>: Self-adaptive LLMs]
 
* 2025-01: [https://arxiv.org/abs/2501.06252 Transformer<sup>2</sup>: Self-adaptive LLMs]
 +
* 2025-08: [https://arxiv.org/abs/2508.14143 Beyond Turing: Memory-Amortized Inference as a Foundation for Cognitive Computation]
  
 
==Parameters as Tokens==
 
==Parameters as Tokens==
Line 114: Line 119:
 
* 2024-10: [https://arxiv.org/abs/2410.01104 softmax is not enough (for sharp out-of-distribution)]
 
* 2024-10: [https://arxiv.org/abs/2410.01104 softmax is not enough (for sharp out-of-distribution)]
 
* 2025-06: [https://arxiv.org/abs/2506.06215 Corrector Sampling in Language Models]
 
* 2025-06: [https://arxiv.org/abs/2506.06215 Corrector Sampling in Language Models]
 +
* 2025-06: [https://arxiv.org/abs/2506.01939 Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning]
 +
* 2025-08: [https://arxiv.org/abs/2508.15260 Deep Think with Confidence] ([https://jiaweizzhao.github.io/deepconf/ project])
  
 
=Missing Elements=
 
=Missing Elements=

Latest revision as of 08:50, 15 September 2025

System 2 Reasoning

See: Increasing AI Intelligence

Memory

LLM Weights Memory

Context Length

Extended Context

Retrieval beyond RAG

See also: AI tools: Retrieval Augmented Generation (RAG)

Working Memory

Long-Term Memory

Storage and Retrieval

Episodic Memory

Updating Weights at Inference-time

Parameters as Tokens

Internal Thought Representation Space

Visual Thinking

Neural (non-token) Latent Representation

Altered Transformer

Tokenization

Generation Order

Diffusion Language Models

Related: Image Synthesis via Autoregression/Diffusion

Sampling

Missing Elements

  • Memory
  • Continuous learning/update
  • Robust contextual model
  • Long-time-horizon coherence
  • Fluid intelligence
  • Agency
  • Modeling of self
  • Daydreaming

Memes

See Also

Retrieved from "http://gisaxs.com/index.php?title=AI_research_trends&oldid=8230"