vLLM V0 to V1: Correctness Before Corrections in RL

Robustness of LLM-Generated SystemVerilog Assertions to Semantics-Preserving RTL Transformations

The paper evaluates how robust large language models are at generating SystemVerilog Assertions (SVA) when the underlying RTL code undergoes semantics‑preserving transformations such as operand reordering, identifier renaming, and redundant parenthesization. Using a curated dataset and two open‑source models (Qwen2.5‑Coder‑7B and DeepSeek‑Coder‑V2‑Lite), the authors find that 9.7%–27.0% of behaviors that were correct on the original RTL become incorrect after transformation, revealing significant instability that aggregate accuracy metrics can hide.

Putting RL back in RLHF

Delay, Plateau, or Collapse: Evaluating the Impact of Systematic Verification Error on RLVR

arXiv:2605. 02909v2 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become a powerful approach for improving the reasoning capabilities of large language models (LLMs).

CASS-RTL: Correctness-Aware Subspace Steering for RTL Generation with LLMs

arXiv:2606. 05680v1 Announce Type: cross Abstract: Recent advances in large language models (LLMs) have enabled the automatic synthesis (generation) of register-transfer level (RTL) code from natural language instructions, offering a promising pathway to accelerate chip design.

Diagnosing Training Inference Mismatch in LLM Reinforcement Learning via a Zero-Mismatch Reference

The paper investigates Training‑Inference Mismatch (TIM) in large‑language‑model reinforcement learning, where rollout generation and policy optimization produce differing token probabilities despite identical model weights. By creating a zero‑mismatch diagnostic setting called VeXact, the authors isolate TIM and demonstrate that even minor token‑level numerical disagreements can trigger training collapse. They further show that TIM alters the effective optimization problem and propose remedies to mitigate its impact.

Gotta Learn Fast: A new benchmark for generalization in RL

Open R1: Update #2

Learning Self-Correction in Vision-Language Models via Rollout Augmentation

arXiv:2602. 08503v2 Announce Type: replace-cross Abstract: Self-correction is essential for solving complex reasoning problems in vision-language models (VLMs).

Geometry-Preserving Orthonormal Initialization for Low-Rank Adaptation in RLVR

arXiv:2606. 31813v1 Announce Type: cross Abstract: Low-rank adaptation (LoRA) and its variants enable parameter-efficient fine-tuning of large language models under the supervised fine-tuning (SFT) paradigm.

LLM4RTL: Tool-Assisted LLM for RTL Generation

arXiv:2606. 15500v1 Announce Type: cross Abstract: Large language models (LLMs) have facilitated impressive progress in software engineering, code generation, tooling, and systems.

Open R1: Update #4

Reinforcing Step-level Reasoning for Effective Self-Correction in LLMs

arXiv:2608. 11573v1 Announce Type: cross Abstract: Achieving effective self-correction, where models verify and correct their own mistakes, remains a fundamental challenge for large language models (LLMs).