Proposes self-function vectors and a controlled evaluation protocol to quantify aleatoric uncertainty in ICL separately from epistemic uncertainty for more reliable LLM confidence measures.
Datasets: A Community Library for Natural Language Processing
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
representative citing papers
Wake Vision pipeline produces a 6M-image person detection dataset for TinyML with 2.2% label error, improving model accuracy up to 6.6% over prior VWW benchmark across architectures and subsets.
Author introduces the ipseome as reusable open infrastructure consisting of the largest free dataset on human identity with public access and versioned files.
Plasticity loss in GPT-style transformers on multilingual tasks persists from 5M to 314M parameters, follows a sublinear scaling law with model size, and occurs in both continual and stationary settings.
StarCLR pretrains on TESS light curves via contrastive learning on overlapping subsequences and improves variable star classification F1 scores over scratch-trained models when fine-tuned on TESS, ZTF, and Gaia.
NV-Embed achieves first place on the MTEB leaderboard across 56 tasks by combining a latent attention layer, causal-mask removal, two-stage contrastive training, and data curation for LLM-based embedding models.
Scaling and instruction tuning increase sycophancy in LLMs on opinion and fact tasks, but a synthetic data fine-tuning intervention reduces it on held-out prompts.
UniD³ applies KG-RAG with Llama 3.3-70B to build six knowledge graphs and generate large validated datasets for drug-disease matching, effectiveness assessment, and target analysis from biomedical literature.
Toeplitz MLP Mixers replace attention with masked Toeplitz multiplications for sub-quadratic complexity while retaining more sequence information and outperforming on copying and in-context tasks.
FedProxy replaces weak adapters with a proxy SLM for federated LLM fine-tuning, outperforming prior methods and approaching centralized performance via compression, heterogeneity-aware aggregation, and training-free fusion.
Module-switching defense disrupts backdoors more effectively than weight averaging with fewer models and remains robust even when some models share the same backdoors.
Multiverse analysis of three published CSS studies reveals substantial variation in findings across methodological decision combinations and identifies cases of computational failure not reported in originals.
citing papers explorer
-
Quantifying Aleatoric Uncertainty of In-Context Learning for Robust Measure of LLM Prediction Confidence
Proposes self-function vectors and a controlled evaluation protocol to quantify aleatoric uncertainty in ICL separately from epistemic uncertainty for more reliable LLM confidence measures.
-
Wake Vision: A Tailored Dataset and Benchmark Suite for TinyML Computer Vision Applications
Wake Vision pipeline produces a 6M-image person detection dataset for TinyML with 2.2% label error, improving model accuracy up to 6.6% over prior VWW benchmark across architectures and subsets.
-
Building the Ipseome: Large, Free, Open, Human Identity Data
Author introduces the ipseome as reusable open infrastructure consisting of the largest free dataset on human identity with public access and versioned files.
-
Can Scale Save Us From Plasticity Loss in Large Language Models?
Plasticity loss in GPT-style transformers on multilingual tasks persists from 5M to 314M parameters, follows a sublinear scaling law with model size, and occurs in both continual and stationary settings.
-
StarCLR: Contrastive Learning Representation for Astronomical Light Curves
StarCLR pretrains on TESS light curves via contrastive learning on overlapping subsequences and improves variable star classification F1 scores over scratch-trained models when fine-tuned on TESS, ZTF, and Gaia.
-
NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models
NV-Embed achieves first place on the MTEB leaderboard across 56 tasks by combining a latent attention layer, causal-mask removal, two-stage contrastive training, and data curation for LLM-based embedding models.
-
Simple synthetic data reduces sycophancy in large language models
Scaling and instruction tuning increase sycophancy in LLMs on opinion and fact tasks, but a synthetic data fine-tuning intervention reduces it on held-out prompts.
-
UniD$^3$: A Knowledge Graph-Enhanced RAG Framework for Drug-Disease Discovery and Reasoning
UniD³ applies KG-RAG with Llama 3.3-70B to build six knowledge graphs and generate large validated datasets for drug-disease matching, effectiveness assessment, and target analysis from biomedical literature.
-
Toeplitz MLP Mixers are Low Complexity, Information-Rich Sequence Models
Toeplitz MLP Mixers replace attention with masked Toeplitz multiplications for sub-quadratic complexity while retaining more sequence information and outperforming on copying and in-context tasks.
-
FedProxy: Federated Fine-Tuning of LLMs via Proxy SLMs and Heterogeneity-Aware Fusion
FedProxy replaces weak adapters with a proxy SLM for federated LLM fine-tuning, outperforming prior methods and approaching centralized performance via compression, heterogeneity-aware aggregation, and training-free fusion.
-
Defending against Backdoor Attacks via Module Switching
Module-switching defense disrupts backdoors more effectively than weight averaging with fewer models and remains robust even when some models share the same backdoors.
-
Making Uncertainty Visible: Multiverse Analysis for Robust Computational Social Science
Multiverse analysis of three published CSS studies reveals substantial variation in findings across methodological decision combinations and identifies cases of computational failure not reported in originals.