Related papers: QSpark: Towards Reliable Qiskit Code Generation

Group-Relative Policy Optimization (GRPO) is a key technique for training large reasoning models, yet it suffers from a critical vulnerability: the \emph{Think-Answer Mismatch}, where noisy reward signals corrupt the learning process. This…

Group Relative Policy Optimization (GRPO) is widely used for critic-free Large Language Model (LLM) post-training, but its KL regularization is usually implemented as a local loss-side token penalty. We show that this misses the…

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a powerful paradigm for reasoning in language models, with GRPO as its primary example. However, GRPO requires continuous online rollout generation, making it…

Online and offline RLHF methods, such as PPO and DPO, have been highly successful in aligning AI with human preferences. Despite their success, however, these methods suffer from fundamental limitations: (a) Models trained with RLHF can…

Quantum signal processing (QSP) provides a systematic framework for implementing a polynomial transformation of a linear operator, and unifies nearly all known quantum algorithms. In parallel, recent works have developed randomized…

High fidelity quantum operations are key to enabling fault-tolerant quantum computation. Superconducting quantum processors have demonstrated high-fidelity operations, but on larger devices there is commonly a broad distribution of…

Quantum machine learning (QML), which combines quantum computing with machine learning, is widely believed to hold the potential to outperform traditional machine learning in the era of noisy intermediate-scale quantum (NISQ). As one of the…

Reinforcement Learning with Verifiable Rewards (RLVR) offers a promising framework for optimizing large language models in reasoning tasks. However, existing RLVR algorithms focus on different granularities, and each has complementary…

Policy optimization methods like Group Relative Policy Optimization (GRPO) and its variants have achieved strong results on mathematical reasoning and code generation tasks. Despite extensive exploration of reward processing strategies and…

Since DeepSeek-R1 popularized, Group Relative Policy Optimization (GRPO) has become the core part of training Reasoning LLMs. However, we find some deficiency that influences RL stability and inference efficiency, like zero-variance in…

Multi-turn tool calling is challenging for Large Language Models (LLMs) because rewards are sparse and exploration is expensive. A common recipe, SFT followed by GRPO, can stall when within-group reward variation is low (e.g., more rollouts…

Group Relative Policy Optimization (GRPO) enhances LLM reasoning but often induces overconfidence, where incorrect responses yield lower perplexity than correct ones, degrading relative calibration as described by the Area Under the Curve…

Layout synthesis, an important step in quantum computing, processes quantum circuits to satisfy device layout constraints. In this paper, we construct QUEKO benchmarks for this problem, which have known optimal depths and gate counts. We…

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an indispensable paradigm for enhancing reasoning in Large Language Models (LLMs). However, standard policy optimization methods, such as Group Relative Policy…

Group Relative Policy Optimization (GRPO) is a standard algorithm for reinforcement learning from verifiable rewards, but its group-mean-centered advantage can fail under binary rewards. The failure mode is gradient starvation: when every…

Group-Relative Policy Optimization (GRPO) has emerged as the standard for training reasoning capabilities in large language models through reinforcement learning. By estimating advantages using group-mean rewards rather than a learned…

Reinforcement learning (RL) has proven remarkably effective at improving the accuracy of language models in verifiable and deterministic domains like mathematics. Here, we examine if current RL methods are also effective at optimizing…

Iterative preference optimization methods have recently been shown to perform well for general instruction tuning tasks, but typically make little improvement on reasoning tasks (Yuan et al., 2024, Chen et al., 2024). In this work we…

Quantum computers show potential for achieving computational advantage over classical computers, with many candidate applications in combinatorial optimisation. We present an application level benchmarking framework for near-term quantum…

Repeated sampling with a verifier is the standard way to allocate test-time compute for code generation, with pass@$K$ as the canonical metric. Yet the standard policy class draws $K$ independent samples from a single answer distribution,…