Large Language Model agents augmented with external skill banks have demonstrated remarkable performance on sequential decision-making tasks. However, existing methods update skills in an undirected, unverified manner and prematurely penalize agents for failing to utilize immature skills. We introduce a structured cognitive representation that decomposes both general and task-specific skills into four dimensions—D-PLAN, D-TRACK, D-EXEC, and D-GUARD—enabling precise failure attribution. Building upon this representation, we propose A²SE, an ability-aligned skill evolution framework that synchronizes skill bank updates with the agent’s learning trajectory. Specifically, A²SE concentrates evolution on the identified bottleneck dimension, safeguards updates via a causal verification mechanism with per-task-type rollback, and employs a competence-aligned skill reward to alleviate premature penalization. Extensive experiments on ALFWorld and WebShop demonstrate that A²SE achieves 11.0% and 7.1% improvements over the strongest baselines, respectively, with ablation studies confirming the necessity of each proposed component.

Long-horizon LLM agents can benefit from reusable skills, yet existing skill-based methods often rely on external skill generators during training or persistent skill retrieval at inference, increasing engineering complexity, context length, and deployment latency. We propose Self-Internalizing Reinforcement learning with Intrinsic skills (SIRI), a three-phase framework that enables agents to discover, validate, and internalize skills without external skill generators or inference-time skill banks. SIRI first warms up the policy with GiGPO to acquire basic interaction ability and collect successful skill-free trajectories. It then performs self-skill mining, where the current policy summarizes compact skills from its own successful plain rollouts and validates them through paired skill-augmented and skill-free rollouts. Finally, SIRI distills only beneficial skill-guided action tokens into the plain policy using trajectory-level utility and action-level advantage. At inference, the agent runs with the original prompt only. On ALFWorld and WebShop with Qwen2.5-7B-Instruct, SIRI improves GiGPO from 0.908 to 0.930 on ALFWorld and from 0.728 to 0.813 on WebShop, outperforming prompt-based, RL-based, and memory-augmented baselines. Further analysis shows that our self-mining strategy can achieve performance comparable to distillation with closed-source large model.

Long-horizon LLM agents require reinforcement learning methods that can assign credit to intermediate decisions under sparse and delayed rewards. Recent group-based methods such as GiGPO improve over trajectory-level GRPO by constructing step-level advantages at repeated anchor states. However, we show that such dense credit can be statistically unreliable: under limited rollouts, rare but lucky actions may receive overly large advantages, producing divergent anchor bias and late-stage training oscillation. We propose Evidence-Calibrated Policy Optimization (ECPO), a critic-free policy optimization algorithm that calibrates step-level credit before policy updates. ECPO combines Evidence-Calibrated Action Advantage, which groups rollouts by canonical actions and shrinks low-count estimates, with Variance-Gated Credit Weighting, which suppresses anchor states dominated by within-action noise. Experiments on ALFWorld and WebShop with Qwen2.5-1.5B/7B show that ECPO consistently outperforms strong baselines, improving GiGPO by +5.2/+7.3 success points on ALFWorld/WebShop with Qwen2.5-1.5B while adding only 0.1% additional advantage-computation overhead.

Machine-generated text (MGT) detection is critical for regulating online information ecosystems, yet existing detectors often fail under limited fresh supervision and humanization-oriented attacks. We study this problem as few-shot robustness recovery under evolving generators, domains, and attack patterns, and use threat modeling to diagnose detector failures from an attacker’s viewpoint. We identify unstable local boundary estimation as a key failure mode: under scarce supervision, humanized MGT can induce prediction flips, logit drift, and clean–adversarial representation misalignment. To address this issue, we propose RAG-GuidEd Attacker Strengthens ConTrastive Few-shot Detector (REACT), a threat-aware few-shot robustness framework. REACT uses a dynamic retrieval-guided attacker to generate boundary-oriented humanized rewrites from extreme positive–negative anchors, and trains the detector with a pairwise boundary contrastive objective to stabilize clean–adversarial local geometry. Alternating attacker–detector updates further form a co-evolutionary curriculum. Experiments on 4 datasets with 4 shot sizes and 3 random seeds show that REACT improves average detection F1 by 4.95 points over 8 SOTA detectors and reduces average attack success rate under 4 strong attacks by 3.66 percentage points.

Machine-generated Text (MGT) detection is crucial for regulating and attributing online texts. While the existing MGT detectors achieve strong performance, they remain vulnerable to simple perturbations and adversarial attacks. To build an effective defense against malicious perturbations, we view MGT detection from a threat modeling perspective, that is, analyzing the model’s vulnerability from an adversary’s point of view and exploring effective mitigations. To this end, we introduce an adversarial framework for training a robust MGT detector, named GREedy Adversary PromoTed DefendER (GREATER). The GREATER consists of two key components: an adversary GREATER-A and a detector GREATER-D. The GREATER-D learns to defend against the adversarial attack from GREATER-A and generalizes the defense to other attacks. GREATER-A identifies and perturbs the critical tokens in embedding space, along with greedy search and pruning to generate stealthy and disruptive adversarial examples. Besides, we update the GREATER-A and GREATER-D synchronously, encouraging the GREATER-D to generalize its defense to different attacks and varying attack intensities. Our experimental results across 10 text perturbation strategies and 6 adversarial attacks show that our GREATER-D reduces the Attack Success Rate (ASR) by 0.67% compared with SOTA defense methods while our GREATER-A is demonstrated to be more effective and efficient than SOTA attack approaches. Codes and dataset are available in https://github.com/Liyuuuu111/GREATER.