SITE applies soft gradient-based head selection to inject ICL-derived task embeddings, outperforming prior embedding adaptation and few-shot ICL across generation, reasoning, and NLU tasks on 12 LLMs from 4B to 70B parameters.
In-context learning creates task vectors
4 Pith papers cite this work, alongside 4 external citations. Polarity classification is still indexing.
4
Pith papers citing it
4
external citations · external index
representative citing papers
Humans and LLMs exhibit similar error patterns in common-sense reasoning, consistent with shared pattern-matching mechanisms rather than abstract world models.
STaR-Quant provides a state-time consistent PTQ framework for DLLMs using SGAT and TAC to improve low-bit weight-activation quantization.
A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.
citing papers explorer
-
Soft Head Selection for Injecting ICL-Derived Task Embeddings
SITE applies soft gradient-based head selection to inject ICL-derived task embeddings, outperforming prior embedding adaptation and few-shot ICL across generation, reasoning, and NLU tasks on 12 LLMs from 4B to 70B parameters.
-
Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
Humans and LLMs exhibit similar error patterns in common-sense reasoning, consistent with shared pattern-matching mechanisms rather than abstract world models.
-
STaR-Quant: State-Time Consistent Post-Training Quantization for Diffusion Large Language Models
STaR-Quant provides a state-time consistent PTQ framework for DLLMs using SGAT and TAC to improve low-bit weight-activation quantization.
-
Correcting Influence: Unboxing LLM Outputs with Orthogonal Latent Spaces
A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.