6 papers · 1 filter
Entropic Context Shaping: Information-Theoretic Filtering for Context-Aware LLM Agents
Hyunjun Kim
Context engineering for large language model (LLM) agents requires distinguishing pragmatically useful information from misleading distractors. We introduce Entropic Context Shapin…
Defensive M2S: Training Guardrail Models on Compressed Multi-turn Conversations
Hyunjun Kim
Guardrail models are essential for ensuring the safety of Large Language Model (LLM) deployments, but processing full multi-turn conversation histories incurs significant computati…
DrawingBench: Evaluating Spatial Reasoning and UI Interaction Capabilities of Large Language Models through Mouse-Based Drawing Tasks
Hyunjun Kim, Sooyoung Ryu
As agentic AI systems increasingly operate autonomously, establishing trust through verifiable evaluation becomes critical. Yet existing benchmarks lack the transparency and audita…
X-Teaming Evolutionary M2S: Automated Discovery of Multi-turn to Single-turn Jailbreak Templates
Hyunjun Kim, Junwoo Ha, Sangyoon Yu +1
Multi-turn-to-single-turn (M2S) compresses iterative red-teaming into one structured prompt, but prior work relied on a handful of manually written templates. We present X-Teaming…
ObjexMT: Objective Extraction and Metacognitive Calibration for LLM-as-a-Judge under Multi-Turn Jailbreaks
Hyunjun Kim, Junwoo Ha, Sangyoon Yu +1
LLM-as-a-Judge (LLMaaJ) enables scalable evaluation, yet we lack a decisive test of a judge's qualification: can it recover the hidden objective of a conversation and know when tha…
M2S: Multi-turn to Single-turn jailbreak in Red Teaming for LLMs
Junwoo Ha, Hyunjun Kim, Sangyoon Yu +4
We introduce a novel framework for consolidating multi-turn adversarial ``jailbreak'' prompts into single-turn queries, significantly reducing the manual overhead required for adve…