paper-with-me

홈 › Papers

BEAVER: An Efficient Deterministic LLM Verifier

2025-12-05 · Tarun Suresh, Nalin Wadhwa, Debangshu Banerjee, Gagandeep Singh arxiv

As large language models (LLMs) transition from research prototypes to production systems, practitioners often need reliable methods to verify model outputs and characterize tail risk for safe deployment. While sampling-based estimates provide an ad-hoc intuition of model behavior, they offer no sound guarantees. We present BEAVER, the first practical framework for computing deterministic, sound probability bounds on LLM satisfaction of safety properties. Given a prompt & any safety property, BEAVER systematically explores the model output space using novel Token trie and Frontier data structures, maintaining provably sound bounds at every iteration. We formalize the verification problem, prove soundness of our approach, and evaluate BEAVER on 4 safety properties across 12 open-weight LLMs. BEAVER identifies 2-3x more risky instances compared to baselines while taking 1/10 of the compute budget, surfacing tail risks that loose bounds and ad-hoc evaluation misses.

📄 PDF Abstract BibTeX arXiv:2512.05439

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

BEAVER: A Training-Free Hierarchical Prompt Compression Method via Structure-Aware Page Selection

2026-03-20 · Zhengpei Hu, Kai Li, Dapeng Fu, Chang Zeng 외 arxiv

The exponential expansion of context windows in LLMs has unlocked capabilities for long-document understanding but introduced severe bottlenecks in inference latency and information utilization. Existing compression meth…

AutoPyVerifier: Learning Compact Executable Verifiers for Large Language Model Outputs

2026-04-24 · Pouya Pezeshkpour, Estevam Hruschka arxiv

Verification is becoming central to both reinforcement-learning-based training and inference-time control of large language models (LLMs). Yet current verifiers face a fundamental trade-off: LLM-based verifiers are expre…

Mathematical Reasoning

BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset

2023-07-10 · NeurIPS 2023 11 · Jiaming Ji, Mickel Liu, Juntao Dai, Xuehai Pan 외

In this paper, we introduce the BeaverTails dataset, aimed at fostering research on safety alignment in large language models (LLMs). This dataset uniquely separates annotations of helpfulness and harmlessness for questi…

Question AnsweringSafety Alignment

PCBSchemaGen: Reward-Guided LLM Code Synthesis for Printed Circuit Boards (PCB) Schematic Design with Structured Verification

2026-01-31 · Huanghaohe Zou, Peng Han, Emad Nazerian, Mafu Zhang 외 arxiv

Most LLM code-synthesis benchmarks rely on unit tests as the reward oracle, but PCB schematic design has none: correctness is defined by structured physical constraints over real IC packages and pin-level assignments, pe…

Provable Defense Against Geometric Transformations

2022-07-22 · Rem Yang, Jacob Laurel, Sasa Misailovic, Gagandeep Singh

Geometric image transformations that arise in the real world, such as scaling and rotation, have been shown to easily deceive deep neural networks (DNNs). Hence, training DNNs to be certifiably robust to these perturbati…

Autonomous DrivingGPU