Adversaries can use crafted scene text to trigger overthinking in LVLM-based robots, producing transferable slowdowns up to 6.96x latency amplification.
InAAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, pages 24312–24320
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
RePoT recovers from PoT failures via deterministic verified replay and checkpoint repair, yielding +3 to +11pp gains on planning benchmarks and showing checkpoint state as the key recovery signal over error-only feedback.
RACER routes between reasoning and non-reasoning LLM judges via constrained distributionally robust optimization to achieve better accuracy-cost trade-offs under distribution shift.
Errors in large reasoning models form a forest structure that grows with more steps, making the first solution best; RED refines the first and prunes the rest for higher performance with less compute.
ACOER applies adaptive correct-only efficiency rewards in GRPO to avoid reward collapse, yielding higher accuracy and over 60% fewer tokens on math reasoning benchmarks.
citing papers explorer
-
Overthink-Triggered Slowdown Attacks on LVLM-Based Robotic Systems
Adversaries can use crafted scene text to trigger overthinking in LVLM-based robots, producing transferable slowdowns up to 6.96x latency amplification.
-
REPOT: Recoverable Program-of-Thought via Checkpoint Repair
RePoT recovers from PoT failures via deterministic verified replay and checkpoint repair, yielding +3 to +11pp gains on planning benchmarks and showing checkpoint state as the key recovery signal over error-only feedback.
-
Reasoning Is Not Free: Robust Adaptive Cost-Efficient Routing for LLM-as-a-Judge
RACER routes between reasoning and non-reasoning LLM judges via constrained distributionally robust optimization to achieve better accuracy-cost trade-offs under distribution shift.
-
FoE: Forest of Errors Makes the First Solution the Best in Large Reasoning Models
Errors in large reasoning models form a forest structure that grows with more steps, making the first solution best; RED refines the first and prunes the rest for higher performance with less compute.
-
Beyond Penalizing Mistakes: Stabilizing Efficiency Training in Large Reasoning Models via Adaptive Correct-Only Rewards
ACOER applies adaptive correct-only efficiency rewards in GRPO to avoid reward collapse, yielding higher accuracy and over 60% fewer tokens on math reasoning benchmarks.