Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents

Large language model (LLM) agents struggle with long-horizon tasks due to their inherent statelessness, requiring all task-relevant information to be encoded in growing input contexts. The resulting degraded reasoning quality, increased inference cost, and higher latency necessitate efficient working memory mechanisms. However, existing approaches either rely on lossy compression or similarity-based retrieval, which often fail to capture temporal structure and causal dependencies required for multi-step agentic tasks. In this work, we present HORMA, a Hierarchical Organize-and-Retrieve Memory Agent that organizes experience into a file-system-like hierarchical structure, where summarized entities are linked to the corresponding raw trajectories, enabling efficient access without losing detailed information. HORMA decomposes working memory into two stages: structured memory construction and navigation-based retrieval. The construction module iteratively refines how experiences are structured by distinguishing between failures caused by missing information and those caused by misleading or overloaded context. The navigation module retrieves task-relevant context by traversing the hierarchy using a lightweight agent trained with reinforcement learning to select minimal yet sufficient context, thereby reducing latency along the critical execution path. Across ALFWorld, LoCoMo, and LongMemEval, HORMA improves task performance under constrained context budgets while requiring at most 22.17% of the baseline token usage in long conversation tasks. Compared to existing methods, it consistently achieves better efficiency-performance trade-offs and generalizes effectively to unseen tasks.

Co-Evolving Skill Generation and Policy Optimization

Skill-augmented reinforcement learning improves language agents by storing reusable procedural knowledge acquired from past experience. Existing methods typically use strong language models to analyze trajectories, generate skills, and update a retrievable skill bank during online training. However, they rarely assess whether a newly generated skill is useful before it is stored and reused. We find that this assumption is unreliable: even skills generated by proprietary frontier LLMs exhibit highly mixed utility, with many providing little benefit or even degrading performance. Once such skills enter the bank, their effects are difficult to identify, because subsequent rollout feedback is delayed and usually reflects the combined effect of multiple retrieved skills rather than the marginal contribution of any individual skill. We propose an online reinforcement learning framework for pre-storage skill validation. The framework estimates whether a candidate skill contributes useful information beyond the skills already retrieved for the current task. It uses the standard rollout budget to form two matched groups under the same task and retrieval context: base rollouts conditioned on the currently retrieved skills, and skill-augmented rollouts conditioned on the same skills plus one candidate skill induced from the base trajectories. The reward gap between these two groups estimates the candidate skill's context-dependent marginal utility, enabling the framework to promote useful skills while filtering ineffective or harmful ones without additional rollout overhead. The framework further uses this marginal-utility signal to train the policy itself as a skill generator, reducing reliance on repeated calls to proprietary models. The learned skill-generation likelihood serves as a context-dependent score for retrieval-time reranking and outdated-skill pruning as the policy evolves.

Learning to Retrieve: Dual-Level Long-Term Memory for Text-to-SQL Agents

Interactive text-to-SQL agents solve database tasks through multi-turn interactions involving schema exploration, query execution, feedback interpretation, and decision revision. Long-term memory helps agents reuse past experiences, but existing retrieval methods remain limited. Static methods rely on fixed similarity heuristics that do not optimize downstream utility, while dynamic methods often learn from sparse final outcomes and retrieve memories at a single decision horizon. This is insufficient when memory usefulness changes across interaction stages, since memories useful for initial planning may differ from those needed for local, state-conditioned execution. We propose MERIT, a dynamic multi-horizon memory retrieval framework. MERIT maintains episode-level memory for global strategic guidance and turn-level memory for local decision support. Both levels use learned retrieval policies optimized with reinforcement learning. To train turn-level retrieval despite limited intermediate supervision, MERIT uses a lightweight Process Reward Model to provide dense proxy rewards for local memory selection. Experiments on BIRD-Interact show that MERIT outperforms no-memory, static-retrieval, and dynamic-retrieval baselines in success rate while reducing average interaction turns. Transfer results on Spider2-Snow further show positive cross-benchmark transfer without benchmark-specific tuning. These results suggest that multi-horizon retrieval improves experience reuse in interactive text-to-SQL agents.

Residual Skill Optimization for Text-to-SQL Ensembles

Text-to-SQL ensembles improve over single-candidate generation by drawing multiple SQL candidates and selecting one, but their effectiveness is bounded by Pass@K, the probability that at least one of K candidates is correct. Existing methods source diversity heuristically through stochastic decoding or prompt variants, leaving candidate sets dominated by correlated failures. We present DivSkill-SQL, a residual skill optimization framework that builds complementary agentic Text-to-SQL ensembles without model fine-tuning: each new skill is optimized on examples the current skill ensemble fails on, provably targeting its marginal contribution to Pass@K. On Spider2-Lite, DivSkill-SQL improves selected accuracy by up to +11.1 points on Snowflake and +8.3 on BigQuery over the strongest ensemble baseline, with consistent gains across two base models (Opus-4.6 and GPT-5.4). Skills optimized on a single dialect transfer without retraining across dialects (Snowflake, BigQuery, SQLite) and to a different task formulation, such as BIRD-Critic (+2.6 pts). Error diagnostics show up to 3x fewer hallucinated schema references and function calls, indicating that gains come from genuinely reliable complementary skills rather than surface-form variation.

Skill-R1: Agent Skill Evolution via Reinforcement Learning

Agentic large language models often rely on skills, reusable natural language procedures that guide planning, action, and tool use. In practice, skills are typically improved through prompt engineering or by aligning the task LLM itself, which is costly, model-specific, and often infeasible for closed-source models. Skill optimization is not a one-step problem but a recurrent process with two coupled levels of credit assignment: a useful skill must improve rollout quality under current conditioning, while a useful revision must turn observed outcomes into a better skill for the next round. We propose Skill-R1, a reinforcement learning framework for instance-level recurrent skill optimization from verifiable rewards. Rather than updating the task LLM, Skill-R1 trains a lightweight skill generator that conditions on the task context, prior rollouts, and their verified outcomes to produce skills that steer a frozen task LLM. This preserves black-box compatibility with both open- and closed-source models while making adaptation substantially cheaper than model-level updates. Skill-R1 proceeds over multiple generations: at each step, the current skill induces rollouts whose verified outcomes are fed back to produce the next revision. To optimize this recurrent process, we introduce a bi-level group-relative policy optimization objective combining intra-generation and inter-generation advantages. The intra-generation term compares rollouts under shared skill conditioning, while the inter-generation term rewards revisions that improve behavior across successive generations. Together, these provide a principled objective for directional skill evolution rather than one-shot self-refinement. Empirically, Skill-R1 achieves consistent gains over no-skill baselines and standard GRPO across benchmarks with verifiable rewards, with particularly strong improvements on complex, multi-step tasks.

Beyond Mode Elicitation: Diversity-Preserving Reinforcement Learning via Latent Diffusion Reasoner

Recent reinforcement learning (RL) methods improve LLM reasoning by optimizing discrete Chain-of-Thought (CoT) generation; however, exploration in token space often suffers from diversity collapse as policy entropy decreases due to mode elicitation behavior in discrete RL. To mitigate this issue, we propose Latent Diffusion Reasoning with Reinforcement Learning (LaDi-RL), a framework that conducts exploration directly in a continuous latent space, where latent variables encode semantic-level reasoning trajectories. By modeling exploration via guided diffusion, multi-step denoising distributes stochasticity and preserves multiple coexisting solution modes without mutual suppression. Furthermore, by decoupling latent-space exploration from text-space generation, we show that latent diffusion-based optimization is more effective than text-space policy optimization alone, while a complementary text policy provides additional gains when combined with latent exploration. Experiments on code generation and mathematical reasoning benchmarks demonstrate consistent improvements in both pass@1 and pass@k over discrete RL baselines, with absolute pass@1 gains of +9.4% on code generation and +5.7% on mathematical reasoning, highlighting diffusion-based latent RL as a principled alternative to discrete token-level RL for reasoning.

LaDiR: Latent Diffusion Enhances LLMs for Text Reasoning

Large Language Models (LLMs) demonstrate their reasoning ability through chain-of-thought (CoT) generation. However, LLM's autoregressive decoding may limit the ability to revisit and refine earlier tokens in a holistic manner, which can also lead to inefficient exploration for diverse solutions. In this paper, we propose LaDiR (Latent Diffusion Reasoner), a novel reasoning framework that unifies the expressiveness of continuous latent representation with the iterative refinement capabilities of latent diffusion models for an existing LLM. We first construct a structured latent reasoning space using a Variational Autoencoder (VAE) that encodes text reasoning steps into blocks of thought tokens, preserving semantic information and interpretability while offering compact but expressive representations. Subsequently, we utilize a latent diffusion model that learns to denoise a block of latent thought tokens with a blockwise bidirectional attention mask, enabling longer horizon and iterative refinement with adaptive test-time compute. This design allows efficient parallel generation of diverse reasoning trajectories, allowing the model to plan and revise the reasoning process holistically. We conduct evaluations on a suite of mathematical reasoning and planning benchmarks. Empirical results show that LaDiR consistently improves accuracy, diversity, and interpretability over existing autoregressive, diffusion-based, and latent reasoning methods, revealing a new paradigm for text reasoning with latent diffusion.

Almost Linear Convergence under Minimal Score Assumptions: Quantized Transition Diffusion

Continuous diffusion models have demonstrated remarkable performance in data generation across various domains, yet their efficiency remains constrained by two critical limitations: (1) the local adjacency structure of the forward Markov process, which restricts long-range transitions in the data space, and (2) inherent biases introduced during the simulation of time-inhomogeneous reverse denoising processes. To address these challenges, we propose Quantized Transition Diffusion (QTD), a novel approach that integrates data quantization with discrete diffusion dynamics. Our method first transforms the continuous data distribution $p_*$ into a discrete one $q_*$ via histogram approximation and binary encoding, enabling efficient representation in a structured discrete latent space. We then design a continuous-time Markov chain (CTMC) with Hamming distance-based transitions as the forward process, which inherently supports long-range movements in the original data space. For reverse-time sampling, we introduce a \textit{truncated uniformization} technique to simulate the reverse CTMC, which can provably provide unbiased generation from $q_*$ under minimal score assumptions. Through a novel KL dynamic analysis of the reverse CTMC, we prove that QTD can generate samples with $O(d\ln^2(d/\epsilon))$ score evaluations in expectation to approximate the $d$--dimensional target distribution $p_*$ within an $\epsilon$ error tolerance. Our method not only establishes state-of-the-art inference efficiency but also advances the theoretical foundations of diffusion-based generative modeling by unifying discrete and continuous diffusion paradigms.

Log-concave Sampling over a Convex Body with a Barrier: a Robust and Unified Dikin Walk

We consider the problem of sampling from a $d$-dimensional log-concave distribution $\pi(\theta) \propto \exp(-f(\theta))$ for $L$-Lipschitz $f$, constrained to a convex body with an efficiently computable self-concordant barrier function, contained in a ball of radius $R$ with a $w$-warm start. We propose a \emph{robust} sampling framework that computes spectral approximations to the Hessian of the barrier functions in each iteration. We prove that for polytopes that are described by $n$ hyperplanes, sampling with the Lee-Sidford barrier function mixes within $\widetilde O((d^2+dL^2R^2)\log(w/\delta))$ steps with a per step cost of $\widetilde O(nd^{\omega-1})$, where $\omega\approx 2.37$ is the fast matrix multiplication exponent. Compared to the prior work of Mangoubi and Vishnoi, our approach gives faster mixing time as we are able to design a generalized soft-threshold Dikin walk beyond log-barrier. We further extend our result to show how to sample from a $d$-dimensional spectrahedron, the constrained set of a semidefinite program, specified by the set $\{x\in \mathbb{R}^d: \sum_{i=1}^d x_i A_i \succeq C \}$ where $A_1,\ldots,A_d, C$ are $n\times n$ real symmetric matrices. We design a walk that mixes in $\widetilde O((nd+dL^2R^2)\log(w/\delta))$ steps with a per iteration cost of $\widetilde O(n^\omega+n^2d^{3\omega-5})$. We improve the mixing time bound of prior best Dikin walk due to Narayanan and Rakhlin that mixes in $\widetilde O((n^2d^3+n^2dL^2R^2)\log(w/\delta))$ steps.

Diff-BBO: Diffusion-Based Inverse Modeling for Black-Box Optimization

Black-box optimization (BBO) aims to optimize an objective function by iteratively querying a black-box oracle. This process demands sample-efficient optimization due to the high computational cost of function evaluations. While prior studies focus on forward approaches to learn surrogates for the unknown objective function, they struggle with high-dimensional inputs where valid inputs form a small subspace (e.g., valid protein sequences), which is common in real-world tasks. Recently, diffusion models have demonstrated impressive capability in learning the high-dimensional data manifold. They have shown promising performance in black-box optimization tasks but only in offline settings. In this work, we propose diffusion-based inverse modeling for black-box optimization (Diff-BBO), the first inverse approach leveraging diffusion models for online BBO problem. Diff-BBO distinguishes itself from forward approaches through the design of acquisition function. Instead of proposing candidates in the design space, Diff-BBO employs a novel acquisition function Uncertainty-aware Exploration (UaE) to propose objective function values, which leverages the uncertainty of a conditional diffusion model to generate samples in the design space. Theoretically, we prove that using UaE leads to optimal optimization outcomes. Empirically, we redesign experiments on the Design-Bench benchmark for online settings and show that Diff-BBO achieves state-of-the-art performance.