Curriculum Learning

Recent advances in collaborative knowledge distillation have demonstrated cutting-edge performance for resource-constrained distributed multimedia learning scenarios. However, achieving such competitiveness requires addressing a fundamental mismatch: high-dimensional teacher knowledge complexity versus heterogeneous client learning capacities, which currently prohibits deployment in edge-based visual analytics systems. Drawing inspiration from curriculum learning principles, we introduce Federated Adaptive Progressive Distillation (FAPD), a consensus-driven framework that orchestrates adaptive knowledge transfer. FAPD hierarchically decomposes teacher features via PCA-based structuring, extracting principal components ordered by variance contribution to establish a natural visual knowledge hierarchy. Clients progressively receive knowledge of increasing complexity through dimension-adaptive projection matrices. Meanwhile, the server monitors network-wide learning stability by tracking global accuracy fluctuations across a temporal consensus window, advancing curriculum dimensionality only when collective consensus emerges. Consequently, FAPD provably adapts knowledge transfer pace while achieving superior convergence over fixed-complexity approaches. Extensive experiments on three datasets validate FAPD's effectiveness: it attains 3.64% accuracy improvement over FedAvg on CIFAR-10, demonstrates 2x faster convergence, and maintains robust performance under extreme data heterogeneity (α=0.1), outperforming baselines by over 4.5%.

Large language model (LLM) agents deployed in unknown environments must learn task structure at test time, but current approaches require thousands of interactions to form useful hypotheses. We present Sensi, an LLM agent architecture for the ARC-AGI-3 game-playing challenge that introduces structured test-time learning through three mechanisms: (1) a two-player architecture separating perception from action, (2) a curriculum-based learning system managed by an external state machine, and (3) a database-as-control-plane that makes the agents context window programmatically steerable. We further introduce an LLM-as-judge component with dynamically generated evaluation rubrics to determine when the agent has learned enough about one topic to advance to the next. We report results across two iterations: Sensi v1 solves 2 game levels using the two-player architecture alone, while Sensi v2 adds curriculum learning and solves 0 levels - but completes its entire learning curriculum in approximately 32 action attempts, achieving 50-94x greater sample efficiency than comparable systems that require 1600-3000 attempts. We precisely diagnose the failure mode as a self-consistent hallucination cascade originating in the perception layer, demonstrating that the architectural bottleneck has shifted from learning efficiency to perceptual grounding - a more tractable problem.

Asymmetric self-play has emerged as a promising paradigm for post-training large language models, where a teacher continually generates questions for a student to solve at the edge of the student's learnability. Although these methods promise open-ended data generation bootstrapped from no human data, they suffer from one major problem: not all problems that are hard to solve are interesting or informative to improve the overall capabilities of the model. Current asymmetric self-play methods are goal-agnostic with no real grounding. We propose Guided Asymmetric Self-Play (GASP), where grounding is provided by real-data goalpost questions that are identified to pose a hard exploration challenge to the model. During self-play, the teacher first generates an easier variant of a hard question, and then a harder variant of that easier question, with the goal of gradually closing the gap to the goalpost throughout training. Doing so, we improve pass@20 on LiveCodeBench (LCB) by 2.5% over unguided asymmetric self-play, and through the curriculum constructed by the teacher, we manage to solve hard goalpost questions that remain out of reach for all baselines.

High-speed legged locomotion struggles with stability and transfer losses at higher command velocities during deployment. One reason is that most curricula vary difficulty along single axis, for example increase the range of command velocities, terrain difficulty, or domain parameters (e.g. friction or payload mass) using either fixed update rule or instantaneous rewards while ignoring how the history of robot training has evolved. We propose TransCurriculum, a transformer-based multi-dimensional curriculum learning approach for agile quadrupedal locomotion. TransCurriculum adapts to 3 axes, velocity command targets, terrain difficulty, and domain randomization parameters (friction and payload mass). Rather than feeding task reward history directly into the low-level control policy, our formulation exploits it at the curriculum level. A transformer-based teacher retrieves the sequence of rewards and uses it to predict future rewards, success rate, and learning progress to guide expansion of this multidimensional curriculum towards high performing task bins. Finally we validate our approach on the Unitree Go1 robot in simulation (Isaac Gym) and deploy it zero-shot on Go1 hardware. Our TransCurriculum policy achieves a maximum velocity of 6.3 m/s in simulation and outperforms prior curriculum baselines. We tested our TransCurriculum trained policy on terrains (carpets, slopes, tiles, concrete), achieving a forward velocity of 4.1 m/s on carpet surpassing the fastest curriculum methods by 18.8% and achieves maximum zero-shot value among all tested methods. Our multi-dimensional curriculum also reduces the transfer loss to 18% from 27% for command only curriculum, demonstrating the benefits of joint training over velocity, terrain and domain randomization dimension while keeping the task success rate of 80-90% on rigid indoor and outdoor surfaces.

Curriculum learning--ordering training examples in a sequence to aid machine learning--takes inspiration from human learning, but has not gained widespread acceptance. Static strategies for scoring item difficulty rely on indirect proxy scores of varying quality and produce curricula that are not specific to the learner at hand. Dynamic approaches base difficulty estimates on gradient information, requiring considerable extra computation during training. We introduce a novel method for measuring the difficulty of individual problem instances that is calibrated to a series of models of increasing competence, and identify transitional problems that are consistently easier as model ability increases. Applying this method to diverse model series constructed from sets of models that are readily available on many tasks, we find that training on a curriculum that levels up from easier to harder transitional problems most efficiently improves a model to the next tier of competence. These problems induce a natural progression from easier to harder items, which outperforms other training strategies. By measuring difficulty directly relative to model competence, our method yields interpretable problems, learner-specific curricula, and a principled basis for step-by-step improvement.

Timestep sampling $p(t)$ is a central design choice in Flow Matching models, yet common practice increasingly favors static middle-biased distributions (e.g., Logit-Normal). We show that this choice induces a speed--quality trade-off: middle-biased sampling accelerates early convergence but yields worse asymptotic fidelity than Uniform sampling. By analyzing per-timestep training losses, we identify a U-shaped difficulty profile with persistent errors near the boundary regimes, implying that under-sampling the endpoints leaves fine details unresolved. Guided by this insight, we propose Curriculum Sampling, a two-phase schedule that begins with middle-biased sampling for rapid structure learning and then switches to Uniform sampling for boundary refinement. On CIFAR-10, Curriculum Sampling improves the best FID from $3.85$ (Uniform) to $3.22$ while reaching peak performance at $100$k rather than $150$k training steps. Our results highlight that timestep sampling should be treated as an evolving curriculum rather than a fixed hyperparameter.

Reinforcement learning with Verifiable Rewards (RLVR) has emerged as a powerful paradigm for eliciting reasoning capabilities in large language models, particularly in mathematics and coding. While recent efforts have extended this paradigm to broader general scientific (STEM) domains, the complex interplay between supervised fine-tuning (SFT) and RL in these contexts remains underexplored. In this paper, we conduct controlled experiments revealing a critical challenge: for general STEM domains, RL applied directly to base models is highly sample-inefficient and is consistently surpassed by supervised fine-tuning (SFT) on moderate-quality responses. Yet sequential SFT followed by RL can further improve performance, suggesting that the two stages play complementary roles, and that how training data is allocated between them matters. Therefore, we propose DeReason, a difficulty-based data decoupling strategy for general reasoning. DeReason partitions training data by reasoning intensity estimated via LLM-based scoring into reasoning-intensive and non-reasoning-intensive subsets. It allocates broad-coverage, non-reasoning-intensive problems to SFT to establish foundational domain knowledge, and reserves a focused subset of difficult problems for RL to cultivate complex reasoning. We demonstrate that this principled decoupling yields better performance than randomly splitting the data for sequential SFT and RL. Extensive experiments on general STEM and mathematical benchmarks demonstrate that our decoupled curriculum training significantly outperforms SFT-only, RL-only, and random-split baselines. Our work provides a systematic study of the interplay between SFT and RL for general reasoning, offering a highly effective and generalized post-training recipe.

Effective exoskeleton assistance requires co-adaptation: as the device alters joint dynamics, the user reorganizes neuromuscular coordination, creating a non-stationary learning problem. Most learning-based approaches do not explicitly account for the sequential nature of human motor adaptation, leading to training instability and poorly timed assistance. We propose Staged Multi-Agent Training (SMAT), a four-stage curriculum designed to mirror how users naturally acclimate to a wearable device. In SMAT, a musculoskeletal human actor and a bilateral hip exoskeleton actor are trained progressively: the human first learns unassisted gait, then adapts to the added device mass; the exoskeleton subsequently learns a positive assistance pattern against a stabilized human policy, and finally both agents co-adapt with full torque capacity and bidirectional feedback. We implement SMAT in the MyoAssist simulation environment using a 26-muscle lower-limb model and an attached hip exoskeleton. Our musculoskeletal simulations demonstrate that the learned exoskeleton control policy produces an average 10.1% reduction in hip muscle activation relative to the no-assist condition. We validated the learned controller in an offline setting using open-source gait data, then deployed it to a physical hip exoskeleton for treadmill experiments with five subjects. The resulting policy delivers consistent assistance and predominantly positive mechanical power without the need for any explicitly imposed timing shift (mean positive power: 13.6 W at 6 Nm RMS torque to 23.8 W at 9.3 Nm RMS torque, with minimal negative power) consistently across all subjects without subject-specific retraining.

Enhancing mathematical reasoning in Large Language Models typically demands massive datasets, yet data efficiency remains a critical bottleneck. While Curriculum Learning attempts to structure this process, standard unidirectional approaches (simple-to-complex) suffer from inefficient sample utilization: they blindly escalate complexity even when foundational gaps persist, leading to wasted computation on unsolvable problems. To maximize the instructional value of every training sample, we introduce a novel Bidirectional Curriculum Generation framework. Unlike rigid trajectories, our multi-agent ecosystem mimics adaptive pedagogy to establish a closed feedback loop. It dynamically generates data by either complicating problems to challenge the model or, crucially, simplying them to repair specific reasoning failures. This mechanism ensures that the model consumes only the most effective data at any given stage. Grounded in the Optimal Pacing Theorem, our approach optimizes the learning trajectory, significantly outperforming baselines while achieving superior reasoning performance with substantially fewer instruction samples.

Target speaker extraction (TSE) aims to isolate a specific speaker's voice from multi-speaker mixtures. Despite strong benchmark results, real-world performance often degrades due to different interacting factors. Previous curriculum learning approaches for TSE typically address these factors separately, failing to capture their complex interactions and relying on predefined difficulty factors that may not align with actual model learning behavior. To address this challenge, we first propose a multi-factor curriculum learning strategy that jointly schedules SNR thresholds, speaker counts, overlap ratios, and synthetic/real proportions, enabling progressive learning from simple to complex scenarios. However, determining optimal scheduling without predefined assumptions remains challenging. We therefore introduce TSE-Datamap, a visualization framework that grounds curriculum design in observed training dynamics by tracking confidence and variability across training epochs. Our analysis reveals three characteristic data regions: (i) easy-to-learn examples where models consistently perform well, (ii) ambiguous examples where models oscillate between alternative predictions, and (iii) hard-to-learn examples where models persistently struggle. Guided by these data-driven insights, our methods improve extraction results over random sampling, with particularly strong gains in challenging multi-speaker scenarios.

Goal-conditioned reinforcement learning has shown considerable potential in robotic manipulation; however, existing approaches remain limited by their reliance on prioritizing collected experience, resulting in suboptimal performance across diverse tasks. Inspired by human learning behaviors, we propose a more comprehensive learning paradigm, ACDC, which integrates multidimensional Adaptive Curriculum (AC) Planning with Dynamic Contrastive (DC) Control to guide the agent along a well-designed learning trajectory. More specifically, at the planning level, the AC component schedules the learning curriculum by dynamically balancing diversity-driven exploration and quality-driven exploitation based on the agent's success rate and training progress. At the control level, the DC component implements the curriculum plan through norm-constrained contrastive learning, enabling magnitude-guided experience selection aligned with the current curriculum focus. Extensive experiments on challenging robotic manipulation tasks demonstrate that ACDC consistently outperforms the state-of-the-art baselines in both sample efficiency and final task success rate.

Post-training large foundation models with reinforcement learning typically relies on massive and heterogeneous datasets, making effective curriculum learning both critical and challenging. In this work, we propose ACTOR-CURATOR, a scalable and fully automated curriculum learning framework for reinforcement learning post-training of large language models (LLMs). ACTOR-CURATOR learns a neural curator that dynamically selects training problems from large problem banks by directly optimizing for expected policy performance improvement. We formulate problem selection as a non-stationary stochastic bandit problem, derive a principled loss function based on online stochastic mirror descent, and establish regret guarantees under partial feedback. Empirically, ACTOR-CURATOR consistently outperforms uniform sampling and strong curriculum baselines across a wide range of challenging reasoning benchmarks, demonstrating improved training stability and efficiency. Notably, it achieves relative gains of 28.6% on AIME2024 and 30.5% on ARC-1D over the strongest baseline and up to 80% speedup. These results suggest that ACTOR-CURATOR is a powerful and practical approach for scalable LLM post-training.

Curriculum learning is a class of training strategies that organizes the data being exposed to a model by difficulty, gradually from simpler to more complex examples. This research explores a reverse curriculum generation approach that recursively decomposes complex datasets into simpler, more learnable components. We propose a teacher-student framework where the teacher is equipped with the ability to reason step-by-step, which is used to recursively generate easier versions of examples, enabling the student model to progressively master difficult tasks. We propose a novel scoring system to measure data difficulty based on its structural complexity and conceptual depth, allowing curriculum construction over decomposed data. Experiments on math datasets (MATH and AIME) and code generation datasets demonstrate that models trained with curricula generated by our approach exhibit superior performance compared to standard training on original datasets.

Fully unsupervised deep generative modeling (FU-DGM) offers significant potential for compressively sampled magnetic resonance imaging (CS-MRI) reconstruction. Representative FU-DGM formulations, such as deep image prior (DIP) and implicit neural representation (INR), employ architectural bias to induce a low-dimensional manifold in the image space that aligns with the forward observation. However, as the underlying inverse system is highly ill-posed, prolonged iterative fitting in FU-DGM typically leads to poor efficiency and noise amplification. In this paper, guided by the cognitive principle of easy-to-hard learning, we propose CogGen, an FU-DGM framework that reformulates CS-MRI reconstruction as a staged inversion problem. Specifically, CogGen implements an self-paced curriculum learning (SPCL)-driven progressive scheduling strategy through an MRI-aware dual-threshold weighting criterion, which adaptively regulates k-space measurement participation. The data-consistency residual thresholding evaluates the fitting reliability of the current generator, while the k-space radius thresholding controls stage-wise measurement exposure, thereby avoiding uniform fitting throughout optimization. Theoretically, our analysis shows that, when early stages favor easy-to-fit measurements, CogGen yields a reduced local sufficient-iteration bound and a smaller cumulative noise-amplification bound, explaining the improved convergence behavior and reconstruction fidelity of CogGen within a finite iteration budget. Numerical experiments demonstrate that both CogGen instantiations, CogGen-DIP and CogGen-INR, achieve superior performance over prevailing CS-MRI reconstruction techniques, including unsupervised and supervised pipelines.

We introduce Predictive Batch Scheduling (PBS), a novel training optimization technique that accelerates language model convergence by dynamically prioritizing high-loss samples during batch construction. Unlike curriculum learning approaches that require predefined difficulty metrics or hard example mining methods that demand expensive per-sample loss tracking, PBS employs a lightweight linear predictor trained online to estimate sample difficulty from static token-level features. Our predictor achieves 0.44 correlation with actual loss using only four simple features: token frequency, sequence length, vocabulary diversity, and rare token ratio. Experiments on a 130M parameter transformer demonstrate that PBS achieves 6-13% faster convergence measured by evaluation loss across training checkpoints, with the predictor's correlation improving from 0.14 to 0.44 over 10,000 training steps. These results validate that token frequency statistics encode meaningful information about sample difficulty, enabling effective curriculum learning with negligible computational overhead.

Large Language Models (LLMs) are changing the coding paradigm, known as vibe coding, yet synthesizing algorithmically sophisticated and robust code still remains a critical challenge. Incentivizing the deep reasoning capabilities of LLMs is essential to overcoming this hurdle. Reinforcement Fine-Tuning (RFT) has emerged as a promising strategy to address this need. However, most existing approaches overlook the heterogeneous difficulty and granularity inherent in test cases, leading to an imbalanced distribution of reward signals and consequently biased gradient updates during training. To address this, we propose Test-driven and cApability-adaptive cuRriculum reinfOrcement fine-Tuning (TAROT). TAROT systematically constructs, for each problem, a four-tier test suite (basic, intermediate, complex, edge), providing a controlled difficulty landscape for curriculum design and evaluation. Crucially, TAROT decouples curriculum progression from raw reward scores, enabling capability-conditioned evaluation and principled selection from a portfolio of curriculum policies rather than incidental test-case difficulty composition. This design fosters stable optimization and more efficient competency acquisition. Extensive experimental results reveal that the optimal curriculum for RFT in code generation is closely tied to a model's inherent capability, with less capable models achieving greater gains with an easy-to-hard progression, whereas more competent models excel under a hard-first curriculum. TAROT provides a reproducible method that adaptively tailors curriculum design to a model's capability, thereby consistently improving the functional correctness and robustness of the generated code. All code and data are released to foster reproducibility and advance community research at https://github.com/deep-diver/TAROT.

Reinforcement learning with verifiable rewards (RLVR) has been a main driver of recent breakthroughs in large reasoning models. Yet it remains a mystery how rewards based solely on final outcomes can help overcome the long-horizon barrier to extended reasoning. To understand this, we develop a theory of the training dynamics of RLVR for transformers on compositional reasoning tasks. Our theory shows that mixed-difficulty training naturally induces an implicit curriculum: without any explicit schedule, easier problems become learnable first and shape the frontier for harder ones, creating a learning progression from easy to hard during optimization. The effectiveness of this curriculum is governed by the smoothness of the difficulty spectrum. When the spectrum is smooth, training dynamics enter a well-behaved relay regime, in which persistent gradient signals on easier problems make slightly harder ones tractable and keep training at the edge of competence. When the spectrum contains abrupt discontinuities, training undergoes grokking-type phase transitions with prolonged plateaus before progress recurs. As a technical contribution, our analysis develops and adapts techniques from Fourier analysis on finite groups to our setting. We validate the predicted mechanisms empirically via controlled synthetic experiments and real-model RLVR runs.

RL methods for scaling large reasoning models stall on datasets with low initial success rates, and thus little training signal. We investigate a fundamental question: Can a pretrained LLM leverage latent knowledge to generate an automated curriculum for problems it cannot solve? We explore this with SOAR: An asymmetric self-play framework that uses meta-RL to surface these pedagogical signals. A teacher model proposes synthetic problems for a student model, and is rewarded with its improvement on a subset of hard problems, thus grounding the curriculum in real student progress rather than intrinsic proxy rewards. Our study on the hardest subsets of math benchmarks (0/128 success) reveals three core findings. First, it is possible to realize bilevel meta-RL that unlocks learning under sparse, binary rewards by sharpening a latent capacity of pretrained models to generate useful problems. Second, grounded rewards outperform intrinsic learnability rewards used in prior LLM self-play, reliably avoiding typical instability and diversity collapse modes. Third, the structure and well-posedness of questions are more critical for learning progress than solution correctness. Our results suggest that the ability to generate useful stepping stones does not require the preexisting ability to solve the hard problems, paving a principled path to escape reasoning plateaus without additional curated data

Large language models (LLMs) increasingly rely on external knowledge to improve factuality, yet many knowledge sources on the Web, such as scholarly indexes and product catalogs, are organized as heterogeneous graphs. Reasoning over such graph-structured knowledge poses two key challenges: (1) navigating structured, schema-defined relations requires precise function calls beyond similarity-based retrieval, and (2) answering complex questions often demands multi-hop evidence aggregation through iterative information seeking. We study how LLMs can learn these interaction capabilities and transfer them to unseen domains. We propose GraphDancer, a reinforcement learning (RL) framework that teaches LLMs to navigate graphs by interleaving reasoning and function execution. To improve RL for moderate-sized LLMs, we introduce a graph-aware curriculum that schedules training by the structural difficulty of questions using an easy-to-hard biased sampler. We evaluate GraphDancer on the multi-domain GRBench benchmark, training only on Academic and Biomedical graphs. Testing covers held-out questions from these domains as well as unseen E-commerce, Literature, and Legal graphs. With either Qwen2.5-3B-Instruct or Qwen3-14B as the backbone, GraphDancer outperforms graph-agent baselines using the same backbone across all evaluation domains. These results indicate that the learned graph exploration and reasoning capabilities transfer across domains. On average across unseen domains, GraphDancer with a 14B backbone also outperforms baselines powered by the closed-source models GPT-4.1-mini and GPT-5-nano. Trajectory analyses further show that GraphDancer retrieves answer-bearing evidence more often and achieves higher accuracy when such evidence is found, while substantially reducing invalid-protocol outcomes.

Quantum neural networks (QNNs) have attracted growing interest for scientific machine learning, yet in regression settings they often suffer from limited trainability under noisy gradients and ill-conditioned optimization. We propose a hybrid quantum--classical regression framework designed to mitigate these bottlenecks. Our model prepends a lightweight classical embedding that acts as a learnable geometric preconditioner, reshaping the input representation to better condition a downstream variational quantum circuit. Building on this architecture, we introduce a curriculum optimization protocol that progressively increases circuit depth and transitions from SPSA-based stochastic exploration to Adam-based gradient fine-tuning. We evaluate the approach on PDE-informed regression benchmarks and standard regression datasets under a fixed training budget in a simulator setting. Empirically, the proposed framework consistently improves over pure QNN baselines and yields more stable convergence in data-limited regimes. We further observe reduced structured errors that are visually correlated with oscillatory components on several scientific benchmarks, suggesting that geometric preconditioning combined with curriculum training is a practical approach for stabilizing quantum regression.

Curriculum learning (CL) structures training from simple to complex samples, facilitating progressive learning. However, existing CL approaches for emotion recognition often rely on heuristic, data-driven, or model-based definitions of sample difficulty, neglecting the difficulty for human perception, a critical factor in subjective tasks like emotion recognition. We propose CHUCKLE (Crowdsourced Human Understanding Curriculum for Knowledge Led Emotion Recognition), a perception-driven CL framework that leverages annotator agreement and alignment in crowd-sourced datasets to define sample difficulty, under the assumption that clips challenging for humans are similarly hard for neural networks. Experimental results suggest that CHUCKLE enhances the performance of LSTMs and Transformers over non-curriculum baselines, while reducing the number of gradient updates, thereby enhancing both training efficiency and model robustness in both subject-dependent and subject-independent settings.

As the cost of pretraining large language models grows, there is continued interest in strategies to improve learning efficiency during this core training stage. Motivated by cognitive development, where humans gradually build knowledge as their brains mature, we propose Curriculum-Guided Layer Scaling (CGLS), a framework for compute-efficient pretraining that synchronizes increasing data difficulty with model growth through progressive layer stacking (i.e. gradually adding layers during training). At the 100M parameter scale, using a curriculum transitioning from synthetic short stories to general web data, CGLS outperforms baseline methods on the question-answering benchmarks PIQA and ARC. Pretraining at the 1.2B scale, we stratify the DataComp-LM corpus with a DistilBERT-based classifier and progress from general text to highly technical or specialized content. Our results show that progressively increasing model depth alongside sample difficulty leads to better generalization and zero-shot performance on various downstream benchmarks. Altogether, our findings demonstrate that CGLS unlocks the potential of progressive stacking, offering a simple yet effective strategy for improving generalization on knowledge-intensive and reasoning tasks.

Curriculum learning enhances Direct Preference Optimization (DPO) for aligning Large Language Models (LLMs), yet existing methods rely on a one-dimensional view of difficulty. In this work, we reframe alignment difficulty as a two-dimensional space spanned by Prompt Complexity (PC) and Pairwise Distinguishability (PD), providing a more principled foundation for alignment. We first demonstrate the efficacy of this space by developing DM-Curri-DPO, a framework of static curricula that already achieves significant gains over baseline methods. Moving beyond these handcrafted paths, we introduce our primary contribution: GSP-Curri-DPO, a novel Group-wise Self-Paced Learning framework. This advanced method empowers the model to navigate the difficulty grid, discovering an optimal learning trajectory based on its own evolving capabilities. Extensive experiments show our self-paced approach not only sets a new state-of-the-art on key benchmarks but, more importantly, demonstrates superior data efficiency and robustness to preference noise. Our work establishes a new paradigm for LLM alignment, offering both a structured difficulty space and an intelligent, model-driven methodology for navigating it.

Reinforcement finetuning (RFT) has shown great potential for enhancing the mathematical reasoning capabilities of large language models (LLMs), but it is often sample- and compute-inefficient, requiring extensive training. In this work, we introduce AdaRFT (Adaptive Curriculum Reinforcement Finetuning), a method that significantly improves the efficiency of RFT through adaptive curriculum learning. AdaRFT dynamically adjusts the difficulty of training problems based on the model's recent reward signals, ensuring that the model consistently trains on tasks that are challenging but solvable. This adaptive sampling strategy accelerates learning by maintaining an optimal difficulty range, avoiding wasted computation on problems that are too easy or too hard. AdaRFT requires only a lightweight extension to standard RFT algorithms like Proximal Policy Optimization (PPO), without modifying the reward function or model architecture. Experiments on competition-level math datasets demonstrate that AdaRFT improves convergence efficiency and reasoning performance. Given problem-level difficulty annotations, AdaRFT reduces RFT training time by up to 2 times across data distributions and model scales, offering a more scalable and effective RFT framework.

Current text-to-3D methods excel at generating single objects but falter on compositional prompts. We argue this failure is fundamental to their optimization schedules, as simultaneous or iterative heuristics predictably collapse under a combinatorial explosion of conflicting gradients, leading to entangled geometry or catastrophic divergence. In this paper, we reframe the core challenge of compositional generation as one of optimization scheduling. We introduce DecompDreamer, a framework built on a novel staged optimization strategy that functions as an implicit curriculum. Our method first establishes a coherent structural scaffold by prioritizing inter-object relationships before shifting to the high-fidelity refinement of individual components. This temporal decoupling of competing objectives provides a robust solution to gradient conflict. Qualitative and quantitative evaluations on diverse compositional prompts demonstrate that DecompDreamer outperforms state-of-the-art methods in fidelity, disentanglement, and spatial coherence.

Amortized inference is the task of training a parametric model, such as a neural network, to approximate a distribution with a given unnormalized density where exact sampling is intractable. When sampling is implemented as a sequential decision-making process, reinforcement learning (RL) methods, such as generative flow networks, can be used to train the sampling policy. Off-policy RL training facilitates the discovery of diverse, high-reward candidates, but existing methods still face challenges in efficient exploration. We propose to use an adaptive training distribution (the \teacher) to guide the training of the primary amortized sampler (the \student). The \teacher, an auxiliary behavior model, is trained to sample high-loss regions of the \student and can generalize across unexplored modes, thereby enhancing mode coverage by providing an efficient training curriculum. We validate the effectiveness of this approach in a synthetic environment designed to present an exploration challenge, two diffusion-based sampling tasks, and four biochemical discovery tasks demonstrating its ability to improve sample efficiency and mode coverage. Source code is available at https://github.com/alstn12088/adaptive-teacher.

A continual learning agent builds on previous experiences to develop increasingly complex behaviors by adapting to non-stationary and dynamic environments while preserving previously acquired knowledge. However, scaling these systems presents significant challenges, particularly in balancing the preservation of previous policies with the adaptation of new ones to current environments. This balance, known as the stability-plasticity dilemma, is especially pronounced in complex multi-agent domains such as the train scheduling problem, where environmental and agent behaviors are constantly changing, and the search space is vast. In this work, we propose addressing these challenges in the train scheduling problem using curriculum learning. We design a curriculum with adjacent skills that build on each other to improve generalization performance. Introducing a curriculum with distinct tasks introduces non-stationarity, which we address by proposing a new algorithm: Continual Deep Q-Network (DQN) Expansion (CDE). Our approach dynamically generates and adjusts Q-function subspaces to handle environmental changes and task requirements. CDE mitigates catastrophic forgetting through EWC while ensuring high plasticity using adaptive rational activation functions. Experimental results demonstrate significant improvements in learning efficiency and adaptability compared to RL baselines and other adapted methods for continual learning, highlighting the potential of our method in managing the stability-plasticity dilemma in the adaptive train scheduling setting.