Presents a new expert-curated dataset of multi-turn counterspeech dialogues in five languages targeting hate against seven groups, with span annotations linking to verified external knowledge for RAG applications.
hub Baseline reference
arXiv preprint arXiv:2311.08370 , year=
Baseline reference. 50% of citing Pith papers use this work as a benchmark or comparison.
14
Pith papers citing it
Baseline
50% of classified citations
hub tools
citation-role summary
dataset 3
background 2
method 1
citation-polarity summary
representative citing papers
DAPRO adaptively spends a fixed interaction budget across multi-turn LLM conversations and builds valid lower bounds on time-to-jailbreak, with coverage error scaling as the square root of the mean censoring weight.
Safety Reflection Pretraining adds regular safety reflections to pretraining data to integrate self-monitoring and reduce unsafe generalization from safe data in LLMs.
Item Response Theory enables adaptive and fixed-subset item selection that reduces safety benchmark costs by 80-99.9% while preserving high correlation with full rankings.
LLMs generate adequate counterspeech for co-occurring hate and misinformation in 40% of cases, with a mixed knowledge strategy from fact-checkers and NGOs proving most effective after expert revision.
Toxicity benchmarks for LLMs produce inconsistent results when task type, input domain, or model changes, revealing intrinsic evaluation biases.
GLiGuard is a compact schema-conditioned bidirectional encoder that matches 7B-27B guard models on safety benchmarks while delivering up to 16x higher throughput and 17x lower latency.
Reasoning traces in large reasoning models expose safety failures missed by final-answer checks, and adaptive multi-principle steering reduces unsafe content in both traces and answers while preserving task performance.
SIREN identifies safety neurons via linear probing on internal LLM layers and combines them with adaptive weighting to detect harm, outperforming prior guard models with 250x fewer parameters.
Language models refuse 75.4% of requests to evade defeated rules and do so even after recognizing reasons that undermine the rule's legitimacy.
WildGuard is a new open moderation model and dataset for LLM safety that identifies harmful prompts, risky responses, and refusal rates, achieving SOTA open-source performance and sometimes exceeding GPT-4 while cutting jailbreak success from 79.8% to 2.4%.
Jailbreak attacks suppress Adversarially Compromised Heads in early layers but leave Safety-Aligned Heads active in mid-layers, producing robust harmful features usable for competitive training-free detection.
Guardian-as-an-Advisor prepends risk labels and explanations from a guardian model to queries, improving LLM safety compliance and reducing over-refusal while adding minimal compute overhead.
Opir introduces efficient multi-task encoder models trained on a 996-category safety taxonomy that match or exceed larger baselines on most safety benchmarks while using under 100M parameters for edge variants.
citing papers explorer
-
CATCH-ME if you RAG: a dataset of Contextually Annotated multi-Turn Counterspeech against Hate and Misinformation Exchanges
Presents a new expert-curated dataset of multi-turn counterspeech dialogues in five languages targeting hate against seven groups, with span annotations linking to verified external knowledge for RAG applications.
-
How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM Evaluation
DAPRO adaptively spends a fixed interaction budget across multi-turn LLM conversations and builds valid lower bounds on time-to-jailbreak, with coverage error scaling as the square root of the mean censoring weight.
-
Beyond Safe Data: Pretraining-Stage Alignment with Regular Safety Reflection
Safety Reflection Pretraining adds regular safety reflections to pretraining data to integrate self-monitoring and reduce unsafe generalization from safe data in LLMs.
-
Efficient Safety Benchmarking via Item Response Theory
Item Response Theory enables adaptive and fixed-subset item selection that reduces safety benchmark costs by 80-99.9% while preserving high correlation with full rankings.
-
Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation
LLMs generate adequate counterspeech for co-occurring hate and misinformation in 40% of cases, with a mixed knowledge strategy from fact-checkers and NGOs proving most effective after expert revision.
-
Navigating the Sea of LLM Evaluation: Investigating Bias in Toxicity Benchmarks
Toxicity benchmarks for LLMs produce inconsistent results when task type, input domain, or model changes, revealing intrinsic evaluation biases.
-
GLiGuard: Schema-Conditioned Classification for LLM Safeguard
GLiGuard is a compact schema-conditioned bidirectional encoder that matches 7B-27B guard models on safety benchmarks while delivering up to 16x higher throughput and 17x lower latency.
-
Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle Steering
Reasoning traces in large reasoning models expose safety failures missed by final-answer checks, and adaptive multi-principle steering reduces unsafe content in both traces and answers while preserving task performance.
-
LLM Safety From Within: Detecting Harmful Content with Internal Representations
SIREN identifies safety neurons via linear probing on internal LLM layers and combines them with adaptive weighting to detect harm, outperforming prior guard models with 250x fewer parameters.
-
Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules
Language models refuse 75.4% of requests to evade defeated rules and do so even after recognizing reasons that undermine the rule's legitimacy.
-
WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs
WildGuard is a new open moderation model and dataset for LLM safety that identifies harmful prompts, risky responses, and refusal rates, achieving SOTA open-source performance and sometimes exceeding GPT-4 while cutting jailbreak success from 79.8% to 2.4%.
-
Robust Harmful Features Under Jailbreak Attacks: Mechanistic Evidence from Attention Head Specialization in Large Language Models
Jailbreak attacks suppress Adversarially Compromised Heads in early layers but leave Safety-Aligned Heads active in mid-layers, producing robust harmful features usable for competitive training-free detection.
-
Guardian-as-an-Advisor: Advancing Next-Generation Guardian Models for Trustworthy LLMs
Guardian-as-an-Advisor prepends risk labels and explanations from a guardian model to queries, improving LLM safety compliance and reducing over-refusal while adding minimal compute overhead.
-
Opir: Efficient Multi-Task Safety Classification for Toxicity, Jailbreaks, Hate Speech, and Harmful Content
Opir introduces efficient multi-task encoder models trained on a 996-category safety taxonomy that match or exceed larger baselines on most safety benchmarks while using under 100M parameters for edge variants.