Pairwise LLM-as-judge evaluation should preserve response identity when presentation order changes. Across four judge models, short A/B prompts yield 70–100% second-position selection (72/80 comparisons overall). Targeted prompt ablations show format sensitivity: label changes alter parsed choices, and a neutral buffer produces complete first-position selection on 59 fixed-order comparisons, reversing rather than removing position preference. We use a position-swapped evaluation protocol, retaining a directional verdict only when both orders select the same response. In a new 16-condition GPT-4.1 factorial audit, analysis-first and a verdict prefix raise directional coverage by 5.6 and 6.5 percentage points, respectively. On 40 Arena-Hard queries with three generators, the 120 pairs comprise 62 directional results, no double ties, 14 swap failures, and 44 invalid pairs. Cross-Model Agreement (CMA) yields 25 unique query winners. A retrospective domain-routing check predicts 17 of these 25 labels using rules fitted outside each test fold. These labels and any oracle score remain judge-relative; human-preference validity and downstream training gains are not established.

Self-distillation for grounded question answering often filters tutor trajectories with agreement or confidence gates. For such gates to support faithful grounding, their signals should depend on the supplied document rather than on answers recoverable from parametric memory. We propose a four-mode randomization-control diagnostic that applies the same frozen gate to original, randomized, irrelevant, and empty documents for the same question. On HotpotQA and MuSiQue, using Qwen2.5-7B-Instruct with eight tutor traces, the joint gate changes sharply with document condition: kept fraction falls from 0.380 to 0.020 and from 0.330 to 0.015, respectively. Randomized and irrelevant documents lie between the endpoints, while a document-blind random gate stays flat. The protocol audits document sensitivity without additional student training; faithful evidence use and downstream gains require separate validation.

Reinforcement learning (RL) is effective for improving code generation but suffers from data scarcity. While experience replay mitigates this, existing approaches rely on static, in-epoch metrics that overlook training dynamics, often introducing low-utility or outdated data. Analyzing RL dynamics via dataset cartography, we observe that “ambiguous” samples, which are vital for model generalization, rapidly migrate to “easy-to-learn” regions, diminishing their training value. To address this, we propose Adaptive Ambiguity Replay (A2R) for RL, a plug-and-play module that prioritizes cross-epoch ambiguous samples. To neutralize the noise from stale experiences, A2R incorporates an adaptive importance mechanism based on policy divergence to weigh replayed rollouts. Extensive experiments on nine LLMs (3B–14B) demonstrate that A2R outperforms state-of-the-art baselines on real-world code editing tasks across both unseen and learned domains. Our results highlight cross-epoch ambiguity as a key factor for effective replay in RL. Code: https://github.com/TsingZ0/verl-A2R

Reinforcement learning (RL) is widely used for post-training large language models (LLMs) in code editing, where group-relative methods, such as GRPO, are popular due to their critic-free and normalized advantage estimation. However, in real-world code-editing scenarios, reward distributions are often skewed with unpredictable noise, leading to distorted advantage computation and increased rollout outliers. To address this issue, we propose Group Adaptive Policy Optimization (GAPO), which adaptively finds an interval with the highest SNR (Signal to Noise Ratio) per prompt and uses the median of that interval as an adaptive Q to replace the group mean in advantage calculation to reduce noise further. This adaptive Q robustly handles rollout noise while remaining plug-and-play and efficient. We evaluate GAPO on nine instruction-tuned LLMs (3B–14B) using a collected large dataset of 51,844 real-world, history-aware code-editing tasks spanning 10 programming languages. GAPO yields up to 4.35 in-domain (ID) and 5.30 out-of-domain (OOD) exact-match improvements over GRPO and its variant DAPO, while achieving lower clipping ratios and higher GPU throughput. Code: https://github.com/TsingZ0/verl-GAPO

Recently, latent reasoning has been introduced into large language models (LLMs) to leverage rich information within a continuous space.However, without stochastic sampling, these methods inevitably collapse to deterministic inference, failing to discover diverse reasoning paths.To bridge the gap, we inject controllable stochasticity into latent reasoning via Gumbel-Softmax, restoring LLMs’ exploratory capacity and enhancing their compatibility with Reinforcement Learning (RL).Building on this, we propose Latent Reasoning Policy Optimization (LEPO), a novel framework that applies RL directly to continuous latent representations.Specifically, in rollout stage, LEPO maintains stochasticity to enable diverse trajectory sampling, while in optimization stage, LEPO constructs a unified gradient estimation for both latent representations and discrete tokens.

Reinforcement Learning with Verifiable Rewards (RLVR) serves as a cornerstone technique for enhancing the reasoning capabilities of Large Language Models (LLMs). However, its training is often plagued by entropy collapse, a rapid decline in policy entropy that limits exploration and undermines training effectiveness. While recent works attempt to mitigate this issue via several heuristic entropy interventions, the underlying mechanisms remain poorly understood. In this work, we conduct comprehensive theoretical and empirical analyses of entropy dynamics in RLVR, offering two main insights: (1) We derive a tight approximation for token-level entropy change at each update step, revealing four governing factors and providing a unified theoretical framework of how existing methods influence entropy; (2) We reveal a fundamental limitation of recent approaches: they rely on heuristic adjustments to one or two of these factors, leaving other relevant factors unconsidered, thus inherently limiting their effectiveness. Motivated by these findings, we propose STEER, a principled entropy-modulation method that adaptively reweighs tokens based on theoretically-estimated entropy variations. Extensive experiments across six mathematical reasoning and three coding benchmarks demonstrate that STEER effectively mitigates entropy collapse and consistently outperforms state-of-the-art baselines.

Large Language Model (LLM) agents are reshaping the industrial landscape. However, most practical agents remain human-designed because tasks differ widely, making them labor-intensive to build. This situation poses a central question: can we automatically create and adapt domain agents in the wild? While several recent approaches have sought to automate agent creation, they typically treat agent generation as a black-box procedure and rely solely on final performance metrics to guide the process. Such strategies overlook critical evidence explaining why an agent succeeds or fails, and often require high computational costs. To address these limitations, we propose ReCreate, an experience-driven framework for the automatic creation of domain agents. ReCreate systematically leverages agent interaction histories, which provide rich concrete signals on both the causes of success or failure and the avenues for improvement. Specifically, we introduce an agent-as-optimizer paradigm that effectively learns from experience via three key components: (i) an experience storage and retrieval mechanism for on-demand inspection; (ii) a reasoning–creating synergy pipeline that map execution experience into scaffold edits; and (iii) hierarchical updates that abstract instance-level details into reusable domain patterns. In experiments across diverse domains, ReCreate consistently outperforms human-designed agents and existing automated agent generation methods, even when starting from minimal seed scaffolds.