AI research trends

System 2 Reasoning

See: Increasing AI Intelligence

Memory

LLM Weights Memory

• 2024-12: Memory Layers at Scale
• 2025-10: Continual Learning via Sparse Memory Finetuning
• 2026-01: Reimagining LLM Memory: Using Context as Training Data Unlocks Models That Learn at Test-Time (Nvidia)
• 2026-01: Entropy-Adaptive Fine-Tuning: Resolving Confident Conflicts to Mitigate Forgetting

Context Length

• 2020: Various ideas for scaling context window, including Longformer
• 2023-04-02: Discussion of ideas for how to scale context window
• 2023-05-11: Anthropic announces 100k window
• 2023-06-07: magic.dev claims 5M tokens coming soon
• 2023-07-05: Microsoft describes LongNet, with 1 billion token window
• 2023-07-11: Focused Transformer 256k
• 2023-11-06: GPT-4 turbo 128k
• 2023-11-22: Anthropic Claude 2.1 200k
• 2023-12-13: Mamba alternative
• 2024-01-04: LongLM to extend context window
• 2024-02-15: Gemini 1.5 1M tokens
• 2024-03-04: Anthropic Claude 3 200k
• 2024-03-08: Google claims Gemini 1.5 can scale to 10M
• 2024-04-10: Google preprint demonstrates infinite context length by using compressive memory
• 2024-04-12: Meta et al. demonstrate Megalodon that enables infinite context via a more efficient architecture
• 2024-04-14: Google presents TransformerFAM, which leverages a feedback loop so it attends to its own latent representations, acting as working memory and provides effectively infinite context
• 2024-10-31: What is Wrong with Perplexity for Long-context Language Modeling?
• 2025-01-14: MiniMax-01 4M (paper)
• 2025-01-27: Qwen2.5-1M (report)
• 2025-02-14: InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU 3M
• 2025-02-18: HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading (code)
• 2025-02-18: Infinite Retrieval: Attention Enhanced LLMs in Long-Context Processing
• 2025-02-19: MoBA: Mixture of Block Attention for Long-Context LLMs
• 2025-02-27: LongRoPE2: Near-Lossless LLM Context Window Scaling (code)
• 2025-03-25: Gemini 2.5 Pro 1M
• 2025-04-05: Meta Llama 4 10M
• 2025-04-14: OpenAI GPT-4.1 1M
• 2025-12-04: Google Titans/Miras 10M
• 2025-12-13: Extending the Context of Pretrained LLMs by Dropping Their Positional Embeddings

Extended Context

• 2025-01: Titans: Learning to Memorize at Test Time

Context Remaking

• 2025-09: ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
• 2025-10: Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models
• 2025-12: Recursive Language Models (model searches/queries the full context)
• 2026-01: SimpleMem: Efficient Lifelong Memory for LLM Agents

Retrieval beyond RAG

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

• 2024-10: Microsoft: KBLaM: Knowledge Base augmented Language Model
• 2024-12: Let your LLM generate a few tokens and you will reduce the need for retrieval
• 2024-12: RetroLLM: Empowering Large Language Models to Retrieve Fine-grained Evidence within Generation
• 2025-03: Microsoft: Introducing KBLaM: Bringing plug-and-play external knowledge to LLMs
• 2025-07: MIRIX: Multi-Agent Memory System for LLM-Based Agents (mirix)
• 2025-08: Memento: Fine-tuning LLM Agents without Fine-tuning LLMs

Working Memory

• 2024-12: Improving Factuality with Explicit Working Memory
• 2026-01: MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory

Long-Term Memory

• 2025-04: Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory

• 2025-12: Google Titans + Miras
  • It's All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimization
  • Titans: Learning to Memorize at Test Time

Storage and Retrieval

• 2025-09: ArcMemo: Abstract Reasoning Composition with Lifelong LLM Memory
• 2026-01: Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

Episodic Memory

• 2024-03: Larimar: Large Language Models with Episodic Memory Control
• 2025-08: AgentFly: Fine-tuning LLM Agents without Fine-tuning LLMs

Continual Learning

• 2025-10: Continual Learning via Sparse Memory Finetuning
• 2025-11: Introducing Nested Learning: A new ML paradigm for continual learning

Updating Weights at Inference-time

• 2025-01: Transformer²: Self-adaptive LLMs
• 2025-08: Beyond Turing: Memory-Amortized Inference as a Foundation for Cognitive Computation

Parameters as Tokens

• 2024-10: TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters (code)

Internal Thought Representation Space

Visual Thinking

• 2025-01: VideoRAG: Retrieval-Augmented Generation over Video Corpus
• 2025-01: Imagine while Reasoning in Space: Multimodal Visualization-of-Thought

Neural (non-token) Latent Representation

• 2024-11: Microsoft: DroidSpeak: KV Cache Sharing for Cross-LLM Communication and Multi-LLM Serving: LLMs invent their own inter-communication language
• 2024-12: Meta: Training Large Language Models to Reason in a Continuous Latent Space: feeding the latent representation directly back into the model, instead of tokenizing intermediate thoughts (Chain of Continuous Thought, a.k.a. Coconut)
• 2024-12: Meta: Large Concept Models: Language Modeling in a Sentence Representation Space: train a model that operates at a higher level of abstraction than typical word/token LLMs; model operates in a space of concept embeddings (more akin to full sentences than individual words)
• 2024-12: Meta: Byte Latent Transformer: Patches Scale Better Than Tokens: Instead of tokenization, dynamically convert input byte-stream into patches, yielding gains in compute efficiency, with minimal loss in performance
• 2024-12: Compressed Chain of Thought: Efficient Reasoning Through Dense Representations
• 2024-12: Google DeepMind: Deliberation in Latent Space via Differentiable Cache Augmentation
• 2024-12: LANG-JEPA: Learning to Think in Latent Space
• 2025-01: Efficient Reasoning with Hidden Thinking (code)
• 2025-02: Huginn-0125: Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach (code, model)
• 2025-02: Meta: LLM Pretraining with Continuous Concepts (CoCoMix)
• 2025-06: Reasoning by Superposition: A Theoretical Perspective on Chain of Continuous Thought

Altered Transformer

Tokenization

• 2024-04: Better & Faster Large Language Models via Multi-token Prediction
• 2024-12: I Don't Know: Explicit Modeling of Uncertainty with an [IDK] Token
• 2025-04: Meta: Multi-Token Attention

Generation Order

• 2019-02: Non-Monotonic Sequential Text Generation
• 2019-04: Mask-Predict: Parallel Decoding of Conditional Masked Language Models
• 2019-06: Sequence Generation: From Both Sides to the Middle
• 2020-04: Probabilistically Masked Language Model Capable of Autoregressive Generation in Arbitrary Word Order
• 2021-12: Spiral Language Modeling
• 2023-10: FiLM: Fill-in Language Models for Any-Order Generation
• 2024-07: Integrating Randomness in Large Language Models: A Linear Congruential Generator Approach for Generating Clinically Relevant Content

Diffusion Language Models

• 2024-02: Pard: Permutation-Invariant Autoregressive Diffusion for Graph Generation
• 2025-02: Large Language Diffusion Models
• 2025-02: Inception Labs Mercury model (online demo)
• 2025-03: Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models (project, code, hf)
• 2025-04: Dream 7B: Introducing Dream 7B, the most powerful open diffusion large language model to date
• 2025-04: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learning (preprint, code)
• 2025-06: Esoteric Language Models (project)

Related: Image Synthesis via Autoregression/Diffusion

• 2023-10: Sequential Data Generation with Groupwise Diffusion Process
• 2024-02: Rolling Diffusion Models
• 2024-08: Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model

Sampling

• 2024-10: entropix: Entropy Based Sampling and Parallel CoT Decoding
• 2024-10: softmax is not enough (for sharp out-of-distribution)
• 2025-06: Corrector Sampling in Language Models
• 2025-06: Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
• 2025-08: Deep Think with Confidence (project)
• 2025-10: Reasoning with Sampling: Your Base Model is Smarter Than You Think

Daydreaming, brainstorming, pre-generation

• Gwern: Daydreaming

Pre-generation

• 2025-03: Reasoning to Learn from Latent Thoughts
• 2025-04: Sleep-time Compute: Beyond Inference Scaling at Test-time
• 2025-11: Project OSSAS: Custom LLMs to process 100 Million Research Papers (models, visualization)

Missing Elements

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

Memes

• Andrej Karpathy:
  • 2015-05: "Hallucination" in The Unreasonable Effectiveness of Recurrent Neural Networks
  • 2017-11: Software 2.0 ("Gradient descent can write code better than you. I'm sorry.")
  • 2022-10: Transformers as general-purpose differentiable computers (talk)
  • 2023-01: The hottest new programming language is English
  • 2023-09: LLM as kernel of a new Operating System (diagram, OS analogies)
  • 2024-07: Jagged Intelligence (c.f. Mollick paper)
  • 2025-02: Vibe coding
  • 2025-06: Software 3.0 (talk): "Prompts as Programs". Software 1.0 is code; 2.0 is model weights; 3.0 is prompts.
  • 2025-06: "Context Engineering" instead of "Prompt Engineering"
  • 2025-06: LLMs as "cognitive cores"
  • 2025-11: Software 1.0 easily automates what you can specify. Software 2.0 easily automates what you can verify.

See Also

• Increasing AI Intelligence