Offline reinforcement learning from static datasets demands expressive policies that remain within data support while improving value. Drifting models train a direct generator via attraction to behavior actions and repulsion among generated particles, but standard drift policies weight repulsion solely by geometric distance and therefore cannot distinguish particles by expected return. This paper proposes Value-Guided Repulsion Q-Learning (VRQL), which modulates repulsion by combining kernel distance with a normalized Q-value bias so that lower-valued neighbors exert stronger repulsive force, while preserving the behavior anchor and a separate value-maximization objective. The authors report competitive performance across 50 OGBench tasks and 6 D4RL AntMaze environments without requiring an auxiliary behavior model.

Federated fine-tuning lets LLMs learn from distributed data, but malicious clients can poison the global model and their attack patterns may evolve as new tasks arrive. This paper proposes Safe-FCL, which continually adapts a risk detector using historical protective subspaces and then performs risk-aware, module-level aggregation of LoRA updates. The goal is to keep the global model safe across sequential tasks without forgetting previously learned attack signatures.

Neural networks require vector inputs, yet many problems involve higher-order objects like probability distributions, functions, and graphs that lack compact representations. This paper proposes a unified encoder-decoder framework that embeds families of such objects into fixed-size vectors using only finite sample sets, with a domain-specific decoder providing the training signal. The core idea is to replace per-object fitting with amortized learning: a permutation-invariant encoder compresses samples into an embedding that conditions a decoder, enabling generalization to unseen family members. The authors instantiate this on distributions, functions, and graphs, and claim it is the first framework general enough to work across object types.

Automated vulnerability repair typically validates patches against the same static alerts or proof-of-concept inputs used to generate them, risking overfitting to partial evidence. This paper introduces P2Patch, a counterexample-guided adversarial repair protocol in which an exploitation agent repeatedly tries to bypass the current patch, with each confirmed bypass becoming a permanent repair requirement. The core idea is to treat the patched program itself as a new attack target and to evaluate final patches against held-out security oracles strictly separated from construction-time evidence.

On-policy self-distillation gives reasoning models dense token-level supervision via a teacher that sees reference solutions, but this signal can drift toward style and ignores whether the student can finish correctly on its own. VASPO repurposes the same privileged reference inside PPO to train both a prefix-value critic and a bounded self-teacher that reweights selected positive advantages. The paper reports average gains of roughly 2.6 percentage points over PPO on math benchmarks and leading accuracy on Knights and Knaves.

Most synthetic pretraining pipelines generate samples by conditioning on an entire source passage or paragraph and asking a model to rewrite or paraphrase it. The paper identifies a systematic failure mode of this approach: fine-grained facts—such as secondary entities, list and table entries, and precise temporal attributes—are frequently omitted because the generator implicitly selects which facts to preserve. Atomic-Fact-Conditioned Synthesis (AFCS) fixes this by first extracting source-supported atomic facts, then conditioning each synthetic training sample on an explicit retained fact and its local context. Continued-pretraining experiments on 7B and 32B models show consistent gains on factual QA benchmarks, especially for middle- and long-tail knowledge, and granularity ablations isolate synthesis granularity as a causal lever for factual recall beyond surface-form diversity.

Probes on hidden activations are a promising tool for detecting deception and harmful intent, but they assume models cannot volitionally alter their own representations. Baldelli et al. ask whether open-weight LLMs can control their activations through in-context neurofeedback, devising a cross-model writer-subject protocol to disentangle genuine self-access from in-context pattern matching. Across 42 semantically opaque readouts and six safety monitors, they find that free-generation effects are largely explained by sentence-selection competence, while fixed-sentence control is tiny and monitor evasion is essentially absent. The results rigorously bound the current threat but flag a capability that may grow with model scale.

CGWM targets the opacity of neural world models by forcing latent states to be explicit symbolic concepts grounded from images by a neural encoder. It replaces learned transition dynamics with hand-specified LTLf rules and performs planning via gradient ascent on fuzzy temporal formulas. The key appeal is interpretability and zero-shot transfer under changing dynamics, though the evaluation is limited to small, fully observable deterministic grids.

Partial-volume (PV) effects blur tissue boundaries in Magnetic Resonance Fingerprinting (MRF), and conventional pipelines rely on multi-stage dictionary matching or statistical segmentation that propagate errors. DeepPV-MRF proposes an end-to-end deep network that directly maps raw, aliased MRF time-series signals to voxel-wise CSF, GM, and WM fraction maps, bypassing intermediate $T_1$/$T_2$ reconstructions. On simulated data the method reports large accuracy gains over dictionary-matching and pseudoinverse baselines, but its real-world evidence rests on synthetic training labels and a small in vivo cohort without ground truth.

Model selection for time-series anomaly detection remains a critical open problem because no single detector dominates across heterogeneous datasets. This paper proposes RAMSAD, a retrieval-based framework that replaces trainable meta-learners with non-parametric similarity search over a knowledge base of past series and their detector-performance profiles. By matching new queries to previously seen instances via time-series foundation model embeddings—optionally augmented with multimodal semantic alignment—the authors argue they can achieve stronger in-distribution accuracy and out-of-distribution robustness at a fraction of the computational cost.

Agent skills let frozen models adapt through editable external instructions, but optimizing them faces a tension: batching many changes obscures which edit mattered, while judging each edit only by its immediate score risks discarding useful stepping stones. AtomSearch proposes contrastive atomic editing—one primitive rule edit tied to execution evidence—and Skill-State Tree Search (STS), which tests short edit compositions and keeps a composed state only if it beats its primitive child. The hope is to find compact, inspectable skills without greedy myopia.

Standard ReAct agents carry every failed attempt and off-path observation in one linear context, which eventually buries the evidence they need. TreeFlex solves this by saving intermediate states as nodes in a search tree and letting the same language model choose where to resume and what to do next, operating in two roles—Reflection for global navigation and Execution for local work—trained end-to-end with a tree-aware adaptation of GRPO. The paper shows that this design improves accuracy on deep research and software engineering benchmarks, especially on harder tasks, though it comes with higher inference costs and small-scale training.

Training large language models is memory-intensive, and subspace projection methods like GaLore can diverge when stochastic noise dominates the selected low-rank gradient space. This paper proposes SMRC and PMRC, which keep optimizer states in a low-dimensional subspace while adding the orthogonal residual gradient back into the parameter update. The authors prove $O(T^{-1/4})$ convergence rates for momentum SGD and for Muon under a mixed stationarity measure, and demonstrate empirical gains on LLM pretraining and fine-tuning up to $1.3$B parameters.

Length-Conditioned Majority Voting (LCMV) improves LLM reasoning by exploiting the empirical bias that longer chain-of-thought traces are less likely to be correct. Instead of filtering out long responses, it learns a logistic prior over correctness conditioned on CoT length and adapts it at test time using only unlabeled data and pseudo-labels from standard majority voting. The method is model-agnostic, adds negligible computational overhead, and is evaluated across 11 models and 8 benchmarks, making it a pragmatic drop-in replacement for standard MV.

The paper studies Protection Mode: a tendency of frontier language models to prioritize protecting a low-risk living subject over an object facing greater immediate physical damage. The authors introduce ProtectionBench, a controlled benchmark of 285 prompts spanning 34 physical scenarios, and find that models frequently choose the subject first even when correctly assessing the object as higher-risk. This matters for understanding how safety alignment and reasoning interact in high-stakes agent decisions.

Existing robot policies use a single fixed parameter configuration across all interaction contexts, which forces shared weights to reconcile competing adaptation signals from diverse manipulation operations. MorphAct addresses this by introducing two coupled generators that synthesize context-dependent low-rank updates for the vision-language and action modules, where the action-side generator is conditioned on the adapted visual-language representations. This allows the effective policy weights to morph at each replanning step without test-time optimization or operation labels. Evaluations across three VLA backbones ($\pi_0$, $\pi_{0.5}$, and X-VLA), two simulation benchmarks, and four real-world tasks report closed-loop success gains of up to 16.30%, accompanied by ablations isolating the value of contextual generation and cross-module coupling.

Long-horizon tasks like real-world software engineering expose the brittleness of monolithic agent loops, where a single model must simultaneously decompose, execute, and verify within one trajectory. LAGO addresses this by treating specialized language agents as temporally extended options—encapsulating role specifications, intra-option policies, termination conditions, and verifiers—that a meta-agent can retrieve from a persistent library or spawn on demand. The core proposition is that this hierarchical scaffold turns task decomposition into a reusable, growing component that requires no weight updates, and the authors report 58.6% on the full 500-task SWE-bench Verified with a frozen Qwen3.5-35B-A3B.

Population dynamics are often observed only as unpaired snapshots, leaving the evolution between measurements fundamentally underdetermined. This paper introduces Neural Wasserstein Splines (NWSP), a method that learns a continuous curve of distributions by penalizing the intrinsic Wasserstein acceleration rather than the acceleration of individual particles, which is representation-dependent. It couples a neural spline generator with a finite-dictionary projection of velocity and its time derivative, achieving competitive results on synthetic and single-cell benchmarks.

Reliable uncertainty quantification for reasoning LLMs is expensive when it requires sampling multiple long chain-of-thought traces, and token-level signals collapse or mis-calibrate on deep reasoning. This paper introduces Argument Graph Uncertainty (AGU), a post-hoc framework for multiple-choice QA that segments a single reasoning trace into propositional nodes, scores pairwise and node-to-option relations with a small NLI model, and aggregates the resulting evidence into a Dirichlet posterior over answer options. The concentration of that posterior, expressed as normalized entropy $U_{\mathrm{AGU}}(R) = H_{\mathrm{norm}}$, yields an uncertainty estimate that is cheap to compute and structurally grounded.

Hindsight-Privileged RLVR tackles the inefficiency of group-based RLVR methods like GRPO, which require multiple rollouts per prompt and lose their learning signal when all sampled outcomes are identical. The core idea is to replace group comparisons with a single rollout per prompt, using the trajectory's own realized future as training-only privilege: a frozen snapshot of the generating policy scores each token under both causal and hindsight views, and normalized Jensen--Shannon divergence between these full-vocabulary distributions weights a historically centered reward. This yields bounded token-level credit without a learned critic or reference solution, training $1.2$--$2.3\times$ faster than GRPO on Qwen3-8B but with task-dependent accuracy tradeoffs.