Dataset Viewer
type
stringclasses 2
values | question
stringlengths 335
1.35k
| instruction
stringclasses 3
values | prompt_template
stringclasses 3
values | doc
sequencelengths 2
11
| answer
stringlengths 1
1.01k
| prompt
stringlengths 165k
1.56M
| chunks
sequencelengths 14
80
|
|---|---|---|---|---|---|---|---|
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["Voice Jailbreak Attacks Against GPT-4o","OR-Bench: An Over-Refusal Benchmark for Large Language Mo(...TRUNCATED) | "{'Reference': ['# CATASTROPHIC JAILBREAK OF OPEN-SOURCE LLMS VIA EXPLOITING GENERATION ', '# Jailbr(...TRUNCATED) | "#Papers:\nVoice Jailbreak Attacks Against GPT-4o\n\nXinyue Shen* Yixin Wu* Michael Backes Yang Zhan(...TRUNCATED) | ["This is part 1 of the document with title: Voice Jailbreak Attacks Against GPT-4o\n\nXinyue Shen* (...TRUNCATED) |
paper | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | {instruction}
#Papers Provided:
{docs} | ["Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language (...TRUNCATED) | "['# Ethical and social risks of harm from Language Models ', '# Using DeepSpeed and Megatron to Tra(...TRUNCATED) | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | ["This is part 1 of the document with title: Using DeepSpeed and Megatron to Train Megatron-Turing N(...TRUNCATED) |
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models","WEAK-TO-STRONG GEN(...TRUNCATED) | "{'Reference': ['# TinyGSM: achieving $>80 \\\\%$ on GSM8k with small language models ', '## WEAK-TO(...TRUNCATED) | "#Papers:\nSelf-Play Fine-Tuning Converts Weak Language Models to Strong Language Models\n\nZixiang (...TRUNCATED) | ["This is part 1 of the document with title: Self-Play Fine-Tuning Converts Weak Language Models to (...TRUNCATED) |
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization","Base of RoPE(...TRUNCATED) | "{'Reference': [], 'Citation': ['# DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Exper(...TRUNCATED) | "#Papers:\nKVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization\n\nCo(...TRUNCATED) | ["This is part 1 of the document with title: KVQuant: Towards 10 Million Context Length LLM Inferenc(...TRUNCATED) |
paper | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | {instruction}
#Papers Provided:
{docs} | ["Minimax Optimal Submodular Optimization with Bandit Feedback","STOCHASTIC SUBMODULAR BANDITS WITH (...TRUNCATED) | "['# Asymptotic Convergence in Online Learning with Unbounded Delays ', '# Stochastic Bandits with D(...TRUNCATED) | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | ["This is part 1 of the document with title: Minimax Optimal Submodular Optimization with Bandit Fee(...TRUNCATED) |
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["Nova ${ }^{+}$: Generative Language Models for Binaries","Generative AI and Large Language Models (...TRUNCATED) | "{'Reference': ['# Nova ${ }^{+}$: Generative Language Models for Binaries '], 'Citation': ['# Gener(...TRUNCATED) | "#Papers:\nNova ${ }^{+}$: Generative Language Models for Binaries\n\nNan Jiang ${ }^{1}$ Chengxiao (...TRUNCATED) | ["This is part 1 of the document with title: Nova ${ }^{+}$: Generative Language Models for Binaries(...TRUNCATED) |
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["A TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decodi(...TRUNCATED) | "{'Reference': ['# PoSE: EFFICIENT CONTEXT WINDOW EXTENSION OF LLMS VIA POSITIONAL SKIP-WISE TRAININ(...TRUNCATED) | "#Papers:\nA TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculati(...TRUNCATED) | ["This is part 1 of the document with title: A TriForce: Lossless Acceleration of Long Sequence Gene(...TRUNCATED) |
paper | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | {instruction}
#Papers Provided:
{docs} | ["Ab initio path integral Monte Carlo simulations of warm dense two-component systems without fixed (...TRUNCATED) | "['# Evidence of free-bound transitions in warm dense matter and their impact on equation-of-state m(...TRUNCATED) | "We kindly ask you to thoroughly review the provided papers and construct a citation chain from them(...TRUNCATED) | ["This is part 1 of the document with title: Ab initio path integral Monte Carlo simulations of warm(...TRUNCATED) |
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","Unbridled I(...TRUNCATED) | "{'Reference': ['# Evil Geniuses: Delving into the Safety of LLM-based Agents ', '# Towards Optimal (...TRUNCATED) | "#Papers:\nRisk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems\n\nT(...TRUNCATED) | ["This is part 1 of the document with title: Risk Taxonomy, Mitigation, and Assessment Benchmarks of(...TRUNCATED) |
paper | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | "We hope you will carefully study the provided papers and determine the citation relationships betwe(...TRUNCATED) | #Papers:
{docs}
{instruction}
#The paper you need to analyze:
{question} | ["Fairness in Large Language Models: A Taxonomic Survey","Risk Taxonomy, Mitigation, and Assessment (...TRUNCATED) | "{'Reference': [], 'Citation': ['# Fairness in Large Language Models: A Taxonomic Survey ', '# Large(...TRUNCATED) | "#Papers:\nFairness in Large Language Models: A Taxonomic Survey\n\nZhibo Chu<br>[email protected](...TRUNCATED) | ["This is part 1 of the document with title: Fairness in Large Language Models: A Taxonomic Survey\n(...TRUNCATED) |
End of preview. Expand
in Data Studio
README.md exists but content is empty.
- Downloads last month
- 38