paper
referenced-only
2021
paper:arxiv-2111-02080An Explanation of In-context Learning as Implicit Bayesian Inference
BySang Michael Xie·Aditi Raghunathan·Percy Liang·Tengyu Ma
Related work— refs + corpus + external arXiv
Cited / in-corpus / arXiv badges show which signals surfaced each row. Multi-source rows weighted higher.
- Implicit Statistical Inference in Transformers: Approximating Likelihood-Ratio Tests In-ContextSiddhant Gadkari Faris Chaudhry2026≈ 79%
- The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and AnalysisJiazheng Li, Yanzheng Xiang, Hanqi Yan, Lin Gui, Yulan He Yuxiang Zhou2024≈ 76%
- ≈ 75%
- Next-token pretraining implies in-context learningPaul M. Riechers and Henry R. Bigelow and Eric A. Alt and Adam Shai2025≈ 74%
- Interpreting Dynamical Systems as Bayesian ReasonersNathaniel Virgo and Martin Biehl and Simon McGregor2021≈ 74%
- A Concise Mathematical Description of Active Inference in Discrete TimeCarlotta Langer, Nihat Ay Jesse van Oostrum2025≈ 74%
- ≈ 74%
- Learning Perception and Planning with Deep Active InferenceTim Verbelen, Johannes Nauta, Cedric De Boom and Bart Dhoedt Ozan \c{C}atal2020≈ 73%
- ≈ 73%
- Active inference for action-unaware agentsKeisuke Suzuki, Ryota Kanai, Manuel Baltieri Filippo Torresan2025≈ 73%
- Active Inference or Control as Inference? A Unifying ViewAbraham Imohiosen, Jan Peters Joe Watson2020≈ 73%
- Demonstrating the Continual Learning Capabilities and Practical Application of Discrete-Time Active InferenceRithvik Prakki2024≈ 73%
- Bayesian model of individual learning to control a motor imagery BCIFabien Lotte (Potioc), J\'er\'emie Mattout C\^ome Annicchiarico2024≈ 73%
- Illusion or Algorithm? Investigating Memorization, Emergence, and Symbolic Processing in In-Context LearningSubhabrata Dutta, Ahmed Elshabrawy, Harish Tayyar Madabushi, Iryna Gurevych Jingcheng Niu2025≈ 73%
- Utility-based Adaptive Teaching Strategies using Bayesian Theory of MindHugo Caselles-Dupr\'e, Olivier Sigaud, Mohamed Chetouani Cl\'emence Grislain2023≈ 72%
- Active inference: demystified and comparedin corpus2021≈ 70%
- ≈ 69%
- ≈ 69%
- Active Inference, Curiosity and Insightin corpus2017≈ 68%
- ≈ 67%
- ≈ 67%
- ≈ 67%
- ≈ 66%
- Active Inference: A Process Theoryin corpus2017≈ 66%
- Learning without neurons in physical systemsin corpus2022≈ 66%
- ≈ 66%
- ≈ 65%
- ≈ 65%
- Emergence and Causality in Complex Systems: A Survey on Causal Emergence and Related Quantitative Studiesin corpus2023≈ 65%
- ≈ 65%
Similar preprints — Semantic Scholar
Cited by (2)
- Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering
In-context learning and activation steering are the same operation on a latent-concept belief: context adds log-likelihood evidence while a steering vector adds a constant shift to the log prior odds,
- The Guanyin Protocol: A Framework for Immediately Establishing an Understanding of Both Causality and Compassion in LLM Systems Using Semantic Anchoring
Semantic anchoring — the binding of a pretrained model's latent patterns to task-specific targets via external structure — predicts threshold-like performance flips with a single calibrated score S =