Emergence of Abstractions: Concept Encoding and Decoding Mechanism for In-Context Learning in Transformers
원문 초록
Humans distill complex experiences into fundamental abstractions that enable rapid learning and adaptation. Similarly, autoregressive transformers exhibit adaptive learning through in-context learning (ICL), which begs the question of how. In this paper, we propose concept encoding-decoding mechanism to explain ICL by studying how transformers form and use internal abstractions in their representations. On synthetic ICL tasks, we analyze the training dynamics of a small transformer and report the coupled emergence of concept encoding and decoding. As the model learns to encode different latent concepts (e.g., ``Finding the first noun in a sentence.") into distinct, separable representations, it concureently builds conditional decoding algorithms and improve its ICL performance. We validate the existence of this mechanism across pretrained models of varying scales (Gemma-2 2B/9B/27B, Llama-3.1 8B/70B). Further, through mechanistic interventions and controlled finetuning, we demonstrate that the quality of concept encoding is causally related and predictive of ICL performance. Our empirical insights shed light into better understanding the success and failure modes of large language models via their representations.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Factors affecting the in-context learning abilities of LLMs for dialogue state tracking
Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression
Mitigating Object Hallucinations via Sentence-Level Early Intervention
Assay2Mol: large language model-based drug design using BioAssay context
AI Wizards at CheckThat! 2025: Enhancing Transformer-Based Embeddings with Sentiment for Subjectivity Detection in News Articles
Dynamic Parameter Memory: Temporary LoRA-Enhanced LLM for Long-Sequence Emotion Recognition in Conversation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.