REVIEW 10 cited by
Trading Inference-Time Compute for Adversarial Robustness
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Trading Inference-Time Compute for Adversarial Robustness
read the original abstract
We conduct experiments on the impact of increasing inference-time compute in reasoning models (specifically OpenAI o1-preview and o1-mini) on their robustness to adversarial attacks. We find that across a variety of attacks, increased inference-time compute leads to improved robustness. In many cases (with important exceptions), the fraction of model samples where the attack succeeds tends to zero as the amount of test-time compute grows. We perform no adversarial training for the tasks we study, and we increase inference-time compute by simply allowing the models to spend more compute on reasoning, independently of the form of attack. Our results suggest that inference-time compute has the potential to improve adversarial robustness for Large Language Models. We also explore new attacks directed at reasoning models, as well as settings where inference-time compute does not improve reliability, and speculate on the reasons for these as well as ways to address them.
Forward citations
Cited by 10 Pith papers
-
From Monoliths to Swarms: A Study of Attack Surface Evolution in the Transition to Multi-Agent Web Systems
A new 'Telephone Loop' attack stalls multi-agent web systems in delegation cycles, succeeding in about 80% of baseline runs for three frontier models while failing against single-agent systems.
-
Addressing Over-Refusal in LLMs with Competing Rewards
SEAR trains one LLM via adversarial process rewards to explore harmful reasoning paths but flip to safe outputs, reducing over-refusal while preserving safety.
-
Chain-of-Models: Cross-Model Auditing for Bias-Robust LLM Judges
Per-bias selection of a cross-family LLM auditor lifts biased-judgment accuracy from 0.805/0.824 baselines to 0.884.
-
Attention-Guided Reward for Reinforcement Learning-based Jailbreak against Large Reasoning Models
An attention-guided RL reward combined with diverse persuasion strategies produces higher attack success rates against large reasoning models than prior jailbreak methods.
-
Controllable LLM Reasoning via Sparse Autoencoder-Based Steering
SAE-Steering finds, via keyword-logit recall plus effectiveness ranking, sparse-autoencoder features that steer a reasoning model into a chosen reasoning strategy, beating baseline steering by ~15% on a judge-based me...
-
Reasoning Up the Instruction Ladder for Controllable Language Models
RLVR on ~7K verifiable system/user conflict examples teaches LLMs to prioritize higher-priority instructions, improving instruction-hierarchy and safety benchmarks.
-
SafeSearch: Automated Red-Teaming of LLM-Based Search Agents
An automated red-teaming framework and 300-case benchmark show that a single unreliable website can induce unsafe responses in LLM search agents, with attack success rates up to 90.5%.
-
Outcome-Guided Distillation: A Teacher-Student Framework to Advance VLM Reasoning in Autonomous Driving
An outcome-guided teacher-student framework for VLM driving, where a teacher generates reflective reasoning from ground-truth waypoints, a student distills it, and a separate decoder converts the reasoning into waypoi...
-
Adaptive Probe-based Steering for Robust LLM Jailbreaking
Adaptive probe-based steering guided by model extraction and activation statistics improves LLM jailbreak success rates from 6% to 70% average harmfulness without extra contrastive prompts or manual tuning.
-
Contrastive Reasoning Alignment: Reinforcement Learning from Hidden Representations
CRAFT uses contrastive representation learning and RL on hidden states to align reasoning models for improved safety against jailbreaks, reporting 79% and 87.7% gains over base models.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.