Search Results for author:
Found 23 papers, 12 papers with code
Why Knowledge Distillation Amplifies Gender Bias and How to Mitigate from the Perspective of DistilBERT
Knowledge distillation is widely used to transfer the language understanding of a large model to a smaller model. However, after knowledge distillation, it was found that the smaller model is more biased by gender compared to the source large model. This paper studies what causes gender bias to increase after the knowledge distillation process. Moreover, we suggest applying a variant of the mixup on knowledge distillation, which is used to increase generalizability during the distillation process, not for augmentation. By doing so, we can significantly reduce the gender bias amplification after knowledge distillation. We also conduct an experiment on the GLUE benchmark to demonstrate that even if the mixup is applied, it does not have a significant adverse effect on the model’s performance.
AdvisorQA: Towards Helpful and Harmless Advice-seeking Question Answering with Collective Intelligence
Therefore, we've completed a benchmark encompassing daily life questions, diverse corresponding responses, and majority vote ranking to train our helpfulness metric.
HyperCLOVA X Technical Report
We introduce HyperCLOVA X, a family of large language models (LLMs) tailored to the Korean language and culture, along with competitive capabilities in English, math, and coding.
Calibrating Large Language Models Using Their Generations Only
As large language models (LLMs) are increasingly deployed in user-facing applications, building trust and maintaining safety by accurately quantifying a model's confidence in its prediction becomes even more important.
KorNAT: LLM Alignment Benchmark for Korean Social Values and Common Knowledge
For the common knowledge dataset, we constructed samples based on Korean textbooks and GED reference materials.
TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identification
Large Language Model (LLM) services and models often come with legal rules on who can use them and how they must use them.
LifeTox: Unveiling Implicit Toxicity in Life Advice
As large language models become increasingly integrated into daily life, detecting implicit toxicity across diverse contexts is crucial.
Prometheus: Inducing Fine-grained Evaluation Capability in Language Models
We first construct the Feedback Collection, a new dataset that consists of 1K fine-grained score rubrics, 20K instructions, and 100K responses and language feedback generated by GPT-4.
KoBBQ: Korean Bias Benchmark for Question Answering
In this paper, we present KoBBQ, a Korean bias benchmark dataset, and we propose a general framework that addresses considerations for cultural adaptation of a dataset.
KoSBi: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Application
Large language models (LLMs) learn not only natural text generation abilities but also social biases against different demographic groups from real-world data.
SQuARe: A Large-Scale Dataset of Sensitive Questions and Acceptable Responses Created Through Human-Machine Collaboration
The potential social harms that large language models pose, such as generating offensive content and reinforcing biases, are steeply rising.
Query-Efficient Black-Box Red Teaming via Bayesian Optimization
To this end, we propose Bayesian red teaming (BRT), novel query-efficient black-box red teaming methods based on Bayesian optimization, which iteratively identify diverse positive test cases leading to model failures by utilizing the pre-defined user input pool and the past evaluations.
Who Wrote this Code? Watermarking for Code Generation
Based on \citet{Kirchenbauer2023watermark}, we propose a new watermarking method, Selective WatErmarking via Entropy Thresholding (SWEET), that promotes "green" tokens only at the position with high entropy of the token distribution during generation, thereby preserving the correctness of the generated code.
Critic-Guided Decoding for Controlled Text Generation
In this work, we propose a novel critic decoding method for controlled language generation (CriticControl) that combines the strengths of reinforcement learning and weighted decoding.
ClaimDiff: Comparing and Contrasting Claims on Contentious Issues
With the growing importance of detecting misinformation, many studies have focused on verifying factual claims by retrieving evidence.
Masked Summarization to Generate Factually Inconsistent Summaries for Improved Factual Consistency Checking
To this end, the latest approach is to train a factual consistency classifier on factually consistent and inconsistent summaries.
Plug-and-Play Adaptation for Continuously-updated QA
To this end, we first propose a novel task--Continuously-updated QA (CuQA)--in which multiple large-scale updates are made to LMs, and the performance is measured with respect to the success in adding and updating knowledge while retaining existing knowledge.
SUMBT+LaRL: Effective Multi-domain End-to-end Neural Task-oriented Dialog System
To our best knowledge, our work is the first comprehensive study of a modularized E2E multi-domain dialog system that learning from each component to the entire dialog policy for task success.
Reasoning Visual Dialog with Sparse Graph Learning and Knowledge Transfer
Visual dialog is a task of answering a sequence of questions grounded in an image using the previous dialog history as context.
SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking
In goal-oriented dialog systems, belief trackers estimate the probability distribution of slot-values at every dialog turn.
Ranked #17 on
Multi-domain Dialogue State Tracking
on MULTIWOZ 2.0
Unpaired Speech Enhancement by Acoustic and Adversarial Supervision for Speech Recognition
Many speech enhancement methods try to learn the relationship between noisy and clean speech, obtained using an acoustic room simulator.
Deep CNNs along the Time Axis with Intermap Pooling for Robustness to Spectral Variations
Convolutional neural networks (CNNs) with convolutional and pooling operations along the frequency axis have been proposed to attain invariance to frequency shifts of features.
Compositional Sentence Representation from Character within Large Context Text
In the HCRN, word representations are built from characters, thus resolving the data-sparsity problem, and inter-sentence dependency is embedded into sentence representation at the level of sentence composition.
