GPT-3 shows that scaling an autoregressive language model to 175 billion parameters enables strong few-shot performance across diverse NLP tasks via in-context prompting without fine-tuning.
On the variance of the adaptive learning rate and beyond
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 4
citation-polarity summary
roles
background 4polarities
background 4representative citing papers
DeBERTa improves BERT-style models by separating content and relative position in attention and adding absolute positions to the decoder, yielding consistent gains on NLU and NLG tasks and the first single-model superhuman score on SuperGLUE.
SmoothLLM mitigates jailbreaking attacks on LLMs by randomly perturbing multiple copies of a prompt at the character level and aggregating the outputs to detect adversarial inputs.
GPTFuzz is a black-box fuzzing framework that mutates seed jailbreak templates to automatically generate effective attacks, achieving over 90% success rates on models including ChatGPT and Llama-2.
HouYi enables prompt injection attacks that grant arbitrary LLM control and steal application prompts in 31 out of 36 tested real-world LLM-integrated applications.
LaMDA shows that fine-tuning on human-value annotations and consulting external knowledge sources significantly improves safety and factual grounding in large dialog models beyond what scaling alone achieves.
Controlled semantic perturbations and selective robustness training with entity masking and adversarial objectives mitigate the typical robustness-accuracy trade-off in publication type and study design classification.
citing papers explorer
-
Language Models are Few-Shot Learners
GPT-3 shows that scaling an autoregressive language model to 175 billion parameters enables strong few-shot performance across diverse NLP tasks via in-context prompting without fine-tuning.
-
DeBERTa: Decoding-enhanced BERT with Disentangled Attention
DeBERTa improves BERT-style models by separating content and relative position in attention and adding absolute positions to the decoder, yielding consistent gains on NLU and NLG tasks and the first single-model superhuman score on SuperGLUE.
-
SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks
SmoothLLM mitigates jailbreaking attacks on LLMs by randomly perturbing multiple copies of a prompt at the character level and aggregating the outputs to detect adversarial inputs.
-
GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts
GPTFuzz is a black-box fuzzing framework that mutates seed jailbreak templates to automatically generate effective attacks, achieving over 90% success rates on models including ChatGPT and Llama-2.
-
Prompt Injection attack against LLM-integrated Applications
HouYi enables prompt injection attacks that grant arbitrary LLM control and steal application prompts in 31 out of 36 tested real-world LLM-integrated applications.
-
LaMDA: Language Models for Dialog Applications
LaMDA shows that fine-tuning on human-value annotations and consulting external knowledge sources significantly improves safety and factual grounding in large dialog models beyond what scaling alone achieves.
-
Robust Biomedical Publication Type and Study Design Classification with Knowledge-Guided Perturbations
Controlled semantic perturbations and selective robustness training with entity masking and adversarial objectives mitigate the typical robustness-accuracy trade-off in publication type and study design classification.