Search Results for author:
Found 14 papers, 6 papers with code
AnaloBench: Benchmarking the Identification of Abstract and Long-context Analogies
Our benchmarking approach focuses on aspects of this ability that are common among humans: (i) recalling related experiences from a large amount of information, and (ii) applying analogical reasoning to complex and lengthy scenarios.
The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts
As the influence of large language models (LLMs) spans across global communities, their safety challenges in multilingual settings become paramount for alignment research.
Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
However, even the top-performing 13B LLM-based translation models, like ALMA, does not match the performance of state-of-the-art conventional encoder-decoder translation models or larger-scale LLMs such as GPT-4.
Narrowing the Gap between Zero- and Few-shot Machine Translation by Matching Styles
Large language models trained primarily in a monolingual setting have demonstrated their ability to generalize to machine translation using zero- and few-shot examples with in-context learning.
Revisiting the Hypothesis: Do pretrained Transformers Learn In-Context by Gradient Descent?
We observe that ICL and GD have different sensitivity to the order in which they observe demonstrations.
SemStamp: A Semantic Watermark with Paraphrastic Robustness for Text Generation
Existing watermarking algorithms are vulnerable to paraphrase attacks because of their token-level design.
The Trickle-down Impact of Reward (In-)consistency on RLHF
We propose Contrast Instructions -- a benchmarking strategy for the consistency of RM.
Sen2Pro: A Probabilistic Perspective to Sentence Embedding from Pre-trained Language Model
Sentence embedding is one of the most fundamental tasks in Natural Language Processing and plays an important role in various tasks.
Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency
We provide theoretical foundations for this metric and its relationship with other prompt selection metrics, providing a comprehensive understanding of existing methods.
Frequency-aware Dimension Selection for Static Word Embedding by Mixed Product Distance
Static word embedding is still useful, particularly for context-unavailable tasks, because in the case of no context available, pre-trained language models often perform worse than static word embeddings.
TextShield: Beyond Successfully Detecting Adversarial Sentences in Text Classification
A recent line of work, detection-based defense, aims to distinguish adversarial sentences from benign ones.
On the Evaluation Metrics for Paraphrase Generation
In this paper we revisit automatic metrics for paraphrase evaluation and obtain two findings that disobey conventional wisdom: (1) Reference-free metrics achieve better performance than their reference-based counterparts.
Rethink the Evaluation for Attack Strength of Backdoor Attacks in Natural Language Processing
It has been shown that natural language processing (NLP) models are vulnerable to a kind of security threat called the Backdoor Attack, which utilizes a `backdoor trigger' paradigm to mislead the models.
