Tagged articles
in-context learning
28 articles · Page 1 of 1
Exin Therapeutics tested GPT-6.1 Sol in three consecutive live mouse brain surgeries where the AI localized Bregma and Lambda; with expert corrections saved as context, the model cut search time from 169 to 86 seconds, corrections from four to two, and localization error from 0.6 to 0.3 millimeters, demonstrating in-context learning from real-world surgical experience.
AI in biomedicineBregmaExin Therapeutics
0 likes · 10 min read

ByteDance Seed team's VINCIE-NExT transfers image editing to video by editing keyframes and using them as visual references, achieving 3.08 on OpenVE-Bench with TDF3D-RoPE positional encoding and Chain-of-Editing inference, though camera viewpoint edits remain challenging.
ByteDance SeedChain-of-EditingOpenVE-Bench
0 likes · 9 min read

Tsinghua researchers propose Jev-LDE, a 1.7B model that performs a single local edit (Keep, Delete, Replace) on retrieved demonstrations for in-context learning, improving average 1-shot classification accuracy from 81.2% to 88.1% across multiple datasets and target LLMs with only 11% inference overhead.
Demonstration SelectionFew-Shot ClassificationGRPO
0 likes · 17 min read

RoboICL enables frozen GPT-6 Astra to control dual-arm robots by providing demonstrations and the robot's own execution history as embodied context, achieving 50.64 average progress score on 30 RoboDojo tasks—surpassing prior methods—with detailed analysis of memory design, real-robot transfer, and 929 executable trajectories for inspection.
Dual-arm ManipulationEmbodied AIGPT-6 Astra
0 likes · 19 min read

The article examines the limits of scaling laws, highlights the pervasive data scarcity in fields such as drug discovery and recommendation, and presents Yaqing Wang’s comprehensive analysis—from few‑shot and meta‑learning to In‑Context Learning and the DEAL framework—demonstrating how structured priors and data‑efficient agentic learning can enable reliable generalization from limited data.
AI for ScienceAgentic LearningData-Efficient Learning
0 likes · 16 min read

The ICML 2026 paper reframes in‑context learning as approximate Bayesian inference, introduces explicit prior datasets as a context prefix for Transformers, and demonstrates through synthetic and real‑world experiments that this multi‑task approach closely matches Bayesian oracles while offering fast, controllable inference.
Bayesian inferenceICML 2026Prior Adaptation
0 likes · 15 min read

The article argues that building AI Agent Skills follows a nascent methodology built on three scientific principles—In‑Context Learning, attention distribution, and bounded rationality—organized into three methodological streams (design‑driven, engineering‑driven, auto‑optimization) and distilled into six reusable design patterns, with a roadmap for future evolution.
AI agentSkill Engineeringautomation
0 likes · 13 min read

The paper introduces a theoretical framework for In‑Context Continual Learning, showing how shared attention in large language models creates bias, variance, and a novel interference term that explains why longer prompts can lead to forgetting, and provides concrete guidelines for prompt design based on task similarity, context length, and order.
Attention MechanismContinual Learningbias-variance-interference
0 likes · 25 min read

A recent study shows that pre‑training Transformers on synthetic, non‑language data generated by Neural Cellular Automata can boost language‑model performance by up to 6%, accelerate convergence by 40%, and improve downstream reasoning, even outperforming models trained on massive natural‑text corpora.
Transformerattention mechanismsin-context learning
0 likes · 12 min read

The article explains how few‑shot prompts—using only 2‑5 task examples—enhance model performance, outlines prompt structure and design principles, examines factors such as example quantity, ordering, label distribution and format, discusses bias sources and calibration, and presents methods for generating many examples with reinforced and unsupervised in‑context learning.
bias mitigationexample designfew-shot prompting
0 likes · 11 min read

The article outlines three post‑RAG knowledge‑engineering approaches—In‑Context Learning with dynamic few‑shot selection, Online Learning encompassing Meta‑Learning and Lifelong Learning to quickly adapt to new tasks, and the Small Language Model path that combines fine‑tuned task‑specific experts with LLM‑SLM collaboration for efficient, privacy‑preserving inference.
Knowledge EngineeringLLMLifelong Learning
0 likes · 4 min read


Amap TechOct 17, 2025 · Artificial Intelligence
This article explains the limitations of current pointwise in‑context learning methods, introduces a novel ranking‑based approach called SeDPO that learns preference orders among examples, and demonstrates its superior performance across multiple NLP tasks through extensive experiments and ablation studies.
NeurIPSRankingSeDPO
0 likes · 10 min read

This article traces the evolution from GPT-1's pre-training and fine-tuning paradigm to GPT-3's in-context learning, contrasting GPT's decoder-only generative approach with BERT's encoder-only bidirectional understanding, and explains zero-shot, one-shot, and few-shot learning.
BERTGPTGenerative Pre-training
0 likes · 7 min read

Prompt engineering uses in‑context learning to turn large language models into precise, task‑aware assistants by providing well‑crafted prompts that guide the model’s probability distribution, reduce hallucinations, and unlock hidden knowledge without any parameter tuning.
artificial intelligencein-context learninglarge language models
0 likes · 6 min read

This survey reviews the evolution of semantic embedding—from Word2vec and GloVe to BERT, Sentence‑BERT, and recent contrastive methods—then examines how large language models improve embeddings via synthetic data generation and backbone architectures, detailing techniques such as contrastive prompting, in‑context learning, knowledge distillation, and discussing resource, privacy, and interpretability challenges.
Contrastive LearningNLPin-context learning
0 likes · 27 min read

FocusICL mitigates the reverse‑scaling of in‑context learning by masking irrelevant tokens and applying hierarchical batch attention, cutting attention complexity, and delivering consistent query focus that yields average accuracy gains of about 5 % across multiple LLMs and benchmarks.
Few-Shot LearningFocusICLattention mechanisms
0 likes · 16 min read

This article reviews the state‑of‑the‑art applications of large language models in recommendation systems, summarizing background knowledge, recent advances such as LLM4Rec, various tuning strategies, agent‑based approaches, open research problems, and future directions for generative recommendation.
AILLMModel tuning
0 likes · 24 min read

At EMNLP 2023, the joint WeChat AI and Peking University paper 'Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning' won the Best Long Paper award, revealing that label tokens act as anchors driving information aggregation in shallow layers and prediction flow in deep layers, and proposing methods to improve and diagnose in‑context learning.
AI researchInformation Flowanchor tokens
0 likes · 13 min read

This article reviews how recent large language models (LLMs) are reshaping recommendation systems, covering their emergence, in‑context learning, prompt‑based strategies, three main LLM‑driven recommendation paradigms, key research papers, experimental results, and future research directions.
Generative AILLMcold start
0 likes · 20 min read


ArchitectApr 19, 2023 · Artificial Intelligence
This article reviews the emergence phenomena observed in large language models, explains how model scale, in‑context learning and chain‑of‑thought prompting contribute to sudden performance gains, discusses small‑model alternatives, and explores the relationship between emergence and the training‑time Grokking effect.
AI researchEmergenceModel Scaling
0 likes · 13 min read

This article recounts a developer's hands‑on experiment using ChatGPT to generate Flutter code for drawing a yellow five‑pointed star, iteratively refining the design, adding explosion particle animation, and reflecting on the strengths and quirks of in‑context learning AI.
AIChatGPTFlutter
0 likes · 13 min read

Prof. Qiu Xipeng’s lecture provides a comprehensive overview of large language models—from their historical development and architectural foundations to key technologies such as in‑context learning, chain‑of‑thought, and natural‑instruction learning, as well as RLHF training, capability evaluation, and current limitations of ChatGPT.
ChatGPTModel EvaluationRLHF
0 likes · 15 min read

This article explains the concept of in‑context learning in large language models, presents experimental evaluations such as copy‑output, date‑formatting, and label‑remapping tasks, and discusses a recent theoretical analysis that links attention layers to implicit gradient‑based fine‑tuning, highlighting why model scale and data volume matter.
Attention MechanismFew-Shot LearningGPT-3
0 likes · 15 min read


ArchitectFeb 18, 2023 · Artificial Intelligence
The article reviews the evolution of large language models, highlighting two major paradigm shifts after GPT‑3, the role of scaling laws, knowledge acquisition, prompting techniques, reasoning abilities, and outlines future research priorities for building more capable and efficient AI systems.
AI ReasoningModel ScalingSparse Transformers
0 likes · 71 min read

In a REDtech live interview, NLP professor Li Lei and Xiaohongshu engineers examined ChatGPT’s strengths—long, topic‑focused replies and few‑shot learning—and its challenges such as hallucinations, safety, lack of real‑time data, model compression, and multimodal AIGC, outlining how the technology could reshape content creation, customer service, and search while requiring careful risk management.
AIAI safetyChatGPT
0 likes · 20 min read

This article compiles three highly upvoted Zhihu answers that examine OpenAI's ChatGPT, discussing its breakthrough impact on NLP, visual in‑context learning, reinforcement‑learning‑from‑human‑feedback, and the broader implications for AI research and development.
AI researchChatGPTRLHF
0 likes · 10 min read


NewBeeNLPFeb 7, 2023 · Artificial Intelligence
This article provides a comprehensive guide to ChatGPT prompt engineering, covering background concepts, design principles, step‑by‑step workflows, diverse use‑case examples, model limitations, and references to key research papers, helping readers craft effective prompts for various NLP tasks.
AIChatGPTNLP
0 likes · 30 min read

The article reviews the evolution of large language models from the pre‑GPT‑3 era to the present, analyzes the conceptual and technical gaps between Chinese and global research, and outlines key future research directions such as scaling laws, prompting techniques, multimodal training, and efficient model architectures.
AI researchChatGPTLLM
0 likes · 73 min read
