RoboScript learns manipulation tasks in context from one demonstration
Authors report 83.4% mean success on all 50 RoboTwin 2.0 tasks without gradient updates, over 20 points above the strongest trained policy.
Stories per day · 30 days
4 RSS
Relatedrobotics2world-models2attention-heads1benchmarks1embodied-ai1gated-associative-lookup1gpt-6-astra1icl-circuit1manipulation1nearest-neighbor1physics-simulation1retrieval1simulation1tabular-classification1transformer-circuits1visual-planning1
31 stories, newest first · one card per story
Stories tagged #in-context-learning
Circuit for In-Context Learning in Language Models
The model implements a gated associative lookup using key-value memory over label-token positions and attention heads for few-shot sentence classification.
PhysiCalWorld calibrates world models from in-context physics clips
One reusable descriptor token, trained only with latent prediction loss, improves rollout accuracy and visual planning.
Study finds nearest-neighbor voting beats LLMs on tabular ICL
Across 19 datasets, Distance-KNN had the highest mean accuracy at every pool size, with LLM gains only under low label consensus.
Authors say GPT-6-Astra has competence, not practical intelligence
Authors report strong benchmark results with in-context learning, but real robots still face costly demos and time limits.
TabWM predicts relational edits from observed executions
Across 512 held-out worlds, complete-edit F1 rose from 0.016 with no executions to 0.685 with 64.
In-Context Learning for System Identification: Probabilistic Formulation, Long-Sequence Handling, and Benchmark Validation
A probabilistic Transformer meta-model performs zero-shot system identification on long sequences with calibrated uncertainty.
Efficient Provably Private Classification with a Tabular Foundation Model
PrivTab is a differentially private tabular foundation model that classifies sensitive data in one forward pass.
GeneICL: A Tabular Foundation Model for Bulk Transcriptomics
GeneICL, a 4.2M-parameter tabular model, leads 80 transcriptomic outcome tasks without inference-time training.
Towards In-Parameter Memory Augmentation for Large Language Models
Survey maps in-parameter memory methods that plug reusable knowledge into LLM inference without growing context.
Recursive Video In-Context Learning for Agentic Robot
RV-ICL lets robot agents navigate one demo video, raising LIBERO success without extra training.
Closing the Context Gap: Activation Alignment for Tabular In-Context Learning
Activation alignment lets tabular models use less context while recovering much of full-context accuracy.
Learning to Learn a Language
A 300M-parameter model trained only on synthetic priors learns to predict real languages in context.
NVIDIA Releases Kumo Tabular: Open Tabular Foundation Models That Predict New Rows in a Single Forward Pass
NVIDIA released Kumo Tabular, open tabular models that predict new rows in one forward pass.
6
Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation
ViGAR plans visual subgoals for robot manipulation, beating the best RoboTwin baseline by 12.86 points.
Effective Dense Retrieval using Only In-Context Examples
RICE turns decoder-only LLMs into training-free dense retrievers by conditioning embeddings on shared in-context examples.
TabFM: A Zero-Shot Foundation Model for Tabular Data
TabFM, a 400M tabular foundation model, delivers calibrated zero-shot predictions that beat tuned AutoML.
In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks
Researchers define robot in-context learning and introduce SimpleICL, a minimal reproducible manipulation framework.
ALICE: In-context, Zero-shot, Mutual Information Estimation
ALICE estimates mutual information zero-shot from samples without retraining on each distribution.
Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It
A rank-8 LoRA lets frozen transformers follow much longer in-context reference chains.
In-Context Learning for Robots: Methods and Applications
A survey groups robot in-context learning into four interfaces from context to action.
Transferring the Intelligence of VLMs to Robotic Control
RoboDawn lets a VLM control robots via discrete commands, beating trained policies zero-shot and one-shot.
Nums AI Releases Causilo: A Tabular Foundation Model That Tops TabArena Among Single Models
Nums AI released Causilo, an Apache-2.0 tabular foundation model achieving the highest single-model Elo (1794) on TabArena for classification and regression.
In-Context Robot Learning with VLM Agents
GPT-Policy uses commercial VLM agents like GPT-6 Astra for in-context robot learning, translating demonstrations and feedback into verified robot actions without gradient updates.