相关论文: CAREF: Calibration-Aware Regularization for Explan…

Large Language Model-based generative recommendation (LLMRec) has achieved notable success, but it suffers from high inference latency due to massive computational overhead and memory pressure of KV Cache. Existing KV Cache reduction…

Consistency regularization-based methods are prevalent in semi-supervised learning (SSL) algorithms due to their exceptional performance. However, they mainly depend on domain-specific data augmentations, which are not usable in domains…

Large language models (LLMs) can acquire new capabilities through fine-tuning, but continual adaptation often leads to catastrophic forgetting. We propose CRAFT, a continual learning framework that avoids updating model weights by instead…

Predictions of uncertainty-aware models are diverse, ranging from single point estimates (often averaged over prediction samples) to predictive distributions, to set-valued or credal-set representations. We propose a novel unified…

Large reasoning models (LRMs) have achieved remarkable success in complex problem-solving, yet they often suffer from computational redundancy or reasoning unfaithfulness. Current methods for shaping LRM behavior typically rely on…

We introduce an Analyze-Revise-Finetune (ARF) pipeline that enables smaller open-source language models (LLMs) to surpass substantially larger proprietary models in customer service summarization tasks. The pipeline first analyzes and…

We introduce a method that dramatically reduces fine-tuning VRAM requirements and rectifies quantization errors in quantized Large Language Models. First, we develop an extremely memory-efficient fine-tuning (EMEF) method for quantized…

Few-shot Continual Event Detection (FCED) poses the dual challenges of learning from limited data and mitigating catastrophic forgetting across sequential tasks. Existing approaches often suffer from severe forgetting due to the full…

Accurate extrinsic calibration of multiple LiDARs is crucial for improving the foundational performance of three-dimensional (3D) map reconstruction systems. This paper presents a novel targetless extrinsic calibration framework for…

Retrieval-augmented large language models, when optimized with outcome-level rewards, can achieve strong answer accuracy on multi-hop questions. However, under noisy retrieval, models frequently suffer from "right-answer-wrong-reason…

Hallucination is a known issue for neural abstractive summarization models. Recent work suggests that the degree of hallucination may depend on errors in the training data. In this work, we propose a new method called Contrastive Parameter…

Calibration measures and reliability diagrams are two fundamental tools for measuring and interpreting the calibration of probabilistic predictors. Calibration measures quantify the degree of miscalibration, and reliability diagrams…

Quantization has become a predominant approach for model compression, enabling deployment of large models trained on GPUs onto smaller form-factor devices for inference. Quantization-aware training (QAT) optimizes model parameters with…

With the rising popularity of Large Language Models (LLMs), there has been an increasing interest in compression techniques that enable their efficient deployment. This study focuses on the Post-Training Quantization (PTQ) of LLMs. Drawing…

We address the problem of network calibration adjusting miscalibrated confidences of deep neural networks. Many approaches to network calibration adopt a regularization-based method that exploits a regularization term to smooth the…

Selective rationales and counterfactual examples have emerged as two effective, complementary classes of interpretability methods for analyzing and training NLP models. However, prior work has not explored how these methods can be…

LLM decoding often relies on the model's predictive distribution to generate an output. Consequently, misalignment with respect to the true generating distribution leads to suboptimal decisions in practice. While a natural solution is to…

As Federated Learning (FL) expands to larger and more distributed environments, consistency in training is challenged by network-induced delays, clock unsynchronicity, and variability in client updates. This combination of factors may…

Sparse autoencoders (SAEs) are widely used to extract human-interpretable features from neural network activations, but their learned features can vary substantially across random seeds and training choices. To improve stability, we studied…

Optimizing vision models purely for classification accuracy can impose an alignment tax, degrading human-like scanpaths and limiting interpretability. We introduce EVA, a neuroscience-inspired hard-attention mechanistic testbed that makes…