Pith. sign in

REVIEW 7 cited by

Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.14160 v4 pith:EMM72R4U submitted 2023-05-23 cs.CL cs.LG

classification cs.CLcs.LG
keywords informationlabeldemonstrationllmsmechanismwordsanchorsexamples
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In-context learning (ICL) emerges as a promising capability of large language models (LLMs) by providing them with demonstration examples to perform diverse tasks. However, the underlying mechanism of how LLMs learn from the provided context remains under-explored. In this paper, we investigate the working mechanism of ICL through an information flow lens. Our findings reveal that label words in the demonstration examples function as anchors: (1) semantic information aggregates into label word representations during the shallow computation layers' processing; (2) the consolidated information in label words serves as a reference for LLMs' final predictions. Based on these insights, we introduce an anchor re-weighting method to improve ICL performance, a demonstration compression technique to expedite inference, and an analysis framework for diagnosing ICL errors in GPT2-XL. The promising applications of our findings again validate the uncovered ICL working mechanism and pave the way for future studies.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Why Multimodal In-Context Learning Lags Behind? Unveiling the Inner Mechanisms and Bottlenecks

    cs.CV 2026-04 unverdicted novelty 7.0 of 10

    Multimodal ICL lags text-only ICL in few-shot settings due to weak cross-modal reasoning alignment and unreliable task mapping transfer, with an inference-stage method proposed to strengthen transfer.

  2. Multimodal Function Vectors for Visual Relations

    cs.AI 2025-10 conditional novelty 6.0 of 10

    Multimodal function vectors extracted from a handful of attention heads in OpenFlamingo-4B encode spatial relations and can be steered, fine-tuned, and composed to improve zero-shot relational reasoning.

  3. Semantic Causality-Aware Vision-Based 3D Occupancy Prediction

    cs.CV 2025-09 conditional novelty 6.0 of 10

    A class-conditional gradient loss (Causal Loss) plus channel-grouped lifting, learnable camera offsets, and normalized convolution raises Occ3D mIoU by 1.2/0.8 points and cuts the camera-noise mIoU drop from 32% to 7%.

  4. PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

    cs.CL 2024-06 conditional novelty 6.0 of 10

    PyramidKV dynamically compresses KV cache across layers following pyramidal information funneling, matching full performance at 12% retention and outperforming alternatives at 0.7% retention with up to 20.5 accuracy gains.

  5. Enhancing Chain-of-Thought Reasoning with Critical Representation Fine-tuning

    cs.CL 2025-07 conditional novelty 5.0 of 10

    CRFT selects critical internal representations via attention and saliency scores and fine-tunes only them, improving GSM8K accuracy over ReFT from 29.0% to 32.8% on LLaMA-2-7B.

  6. NEAT: Concept driven Neuron Attribution in LLMs

    cs.CL 2025-08 reject novelty 4.0 of 10

    NEAT identifies concept neurons by feeding a single mean hidden-state vector through the model and ranking neurons by their effect on concept-word probabilities.

  7. MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Models

    cs.CV 2025-07 conditional novelty 4.0 of 10

    MCA-LLaVA reindexes image tokens by sums of mirrored 2D coordinates so instruction tokens attend across the whole image, reducing hallucination on POPE, CHAIR, and MME.

Pith tools