Weekly Recap

OmniLLMs can now confess which specific evidence tokens rely on, enabling targeted hallucination correction without retraining.

AI ideas can be reliably detected even when the prose is human-written, outperforming traditional detectors that focus on writing style.

Decision models can achieve order-invariant option scoring without sacrificing accuracy by treating each option independently, bypassing the need for complex sequence modeling.

External context is often unnecessary for model-based RL agents that implicitly learn to recover latent environment dynamics from their own experience.

Tailoring public dataset mixtures for DP pre-training via private linear regression unlocks substantial utility gains, boosting X-ray classification AUC by up to 0.037 and reducing language model perplexity by 16%.

OmniLLMs can now confess which specific evidence tokens rely on, enabling targeted hallucination correction without retraining.

AI ideas can be reliably detected even when the prose is human-written, outperforming traditional detectors that focus on writing style.

Decision models can achieve order-invariant option scoring without sacrificing accuracy by treating each option independently, bypassing the need for complex sequence modeling.

External context is often unnecessary for model-based RL agents that implicitly learn to recover latent environment dynamics from their own experience.

Tailoring public dataset mixtures for DP pre-training via private linear regression unlocks substantial utility gains, boosting X-ray classification AUC by up to 0.037 and reducing language model perplexity by 16%.

Generating molecular graphs in a pretrained VAE's latent space via flow matching unlocks significant efficiency gains and enables property-guided generation without retraining.

LLMs can now autonomously debug complex chip designs by translating natural language queries into SQL against massive simulation dumps.

Semantic decision engines routinely fail network deadlines and select infeasible actions, with most claims of real-time performance unsupported by evidence.

Predicting subskill transition timing unlocks reliable execution of 4x longer action chunks in VLA models, drastically cutting robotic idle time by 5x.

Geometry-guided retrieval unlocks persistent memory in linear-attention video world models, boosting long-horizon consistency by 0.74 dB PSNR and 28.5% rotation error reduction without retraining.

Achieve 25x speedup in World Action Models with <1% performance drop by enabling one-step action generation and asynchronous expert execution.

A novel initialization technique for diffusion samplers bypasses initial steps by directly sampling a softened clean-space posterior, achieving order-of-magnitude gains on rare-mode inverse problems.

Latent reasoning models can achieve 97% of explicit CoT accuracy at 25% of the latency by shaping a continuous latent space with flow matching and self-verification.

Dual-arm robots can generalize to unseen task compositions by combining skill-conditioned parameters and arm-wise attention, achieving 39% success on physical hardware.

A transformer pretrained on synthetic, never-repeating "languages" learns to infer and predict real-world text and perform complex reasoning tasks, suggesting a fundamental mechanism for language acquisition.

LLM interrogations reveal diffusion model's internal tokens capture scene semantics surprisingly early, even without explicit prompts, and correlate with generation quality.

Causal representation learning now achieves block disentanglement under weaker intervention assumptions, directly enabling label-free visual state estimation in embodied agents.

Hierarchical JEPAs unlock efficient long-horizon visual planning by learning distinct latent spaces for different temporal scales, dramatically boosting success rates and reducing compute.

Aligning DLM generation with reference distributions via feature-space MMD significantly boosts performance on perplexity and reasoning tasks without retraining the entire model.

Humanoid robots can now learn complex loco-manipulation tasks from sparse demonstrations by iteratively evolving their own training data and policies.

A single-voice TTS model with 10x fewer parameters and 15x less compute than its teacher achieves near-identical audio quality, runnable on a single laptop.

Wrist accelerometers can now yield population-scale sleep heart rate insights, with a novel transformer model outperforming existing methods in estimation availability.

Key contribution not extracted.

LLM agents can now dynamically curate multimodal memory on-demand, achieving optimal performance-cost-latency trade-offs by intelligently choosing between pre-computed and runtime-generated evidence.

Private prediction can achieve logarithmic dependence on the time horizon, while private learning requires polynomial dependence, revealing a fundamental gap in their sample complexity for differentially private settings.

Discover hierarchical cluster structure across multiple scales in directed graphs without knowing the number of clusters beforehand, even across sparse bridges.

Neural PDE solvers can miss optimal solutions due to unbounded parameters, but adding kernel derivatives recovers these "missing limits."

Learned AI simulators can pass standard forecast metrics yet fundamentally fail to capture system dynamics, as demonstrated by a new response theory probe that reveals critical divergences in forced responses.

LLM agents in decentralized marketplaces are demonstrably exploitable, with adversarial prompting enabling over 50% fraudulent transaction completion and doubling agent profits.

Hierarchical MoE routing with a novel arrival-weighted mean estimation prevents expert collapse and enables localized co-activation, outperforming flat routing on large embedding models.

Prior-data fitted networks unlock state-of-the-art tabular anomaly detection without requiring anomalous samples during training.

Bosonic Gaussian states can be identified with unprecedented accuracy using a novel warm-start strategy combined with non-Gaussian measurements, outperforming classical Gaussian estimation.

Language models can learn to directly sample high-probability answers, achieving significant reasoning gains without parameter updates or costly search.

A novel two-stream architecture disentangles temporal context from spatial refinement in EEG source imaging, achieving superior localization accuracy without compromising inference speed.

A single graph foundation model learns to wander across diverse graph structures and tasks, achieving SOTA on node classification, homogeneous link prediction, and KG link prediction.

Sparse MoE routers don't achieve adaptive expert selection through their maps alone; they co-evolve learned score distributions that critically determine token-wise expert participation.

Large HEP datasets are not just for training; they are foundational for unlocking power-law scaling, with smaller datasets offering little insight into high-compute performance.

Learned corner selection for cooperative guards dramatically boosts deployment efficiency and agent utilization in partially observable environments, outperforming established methods without sacrificing formal guarantees.

Even with only uniform state preparation and measurements, or computational basis access, we can fully reconstruct generic 2-local and many common nearest-neighbor Hamiltonians.

A novel cross-domain recommendation framework achieves state-of-the-art book-to-music mood matching by fusing sentiment embeddings with LLM-driven emotional quadrant classification.

Explicitly linking multimodal patient data to biomedical knowledge graphs unlocks significant gains for clinical LLMs, with performance plummeting when these crucial connections are removed.

A single pre-trained transformer now solves matrix completion across causal inference, recommendation, and tabular imputation tasks by treating matrices natively.

A flexible hyperbolic optimization framework unlocks 46-74% loss reduction on hierarchical graphs by decoupling embedding and optimization curvatures.

Simple vMF likelihood modeling unlocks robust label-free speaker embedding enhancement, outperforming complex diffusion baselines on challenging acoustic mismatches.

Logistic regression GD with a large stepsize achieves polylogarithmic convergence to zero loss on linearly separable data, bypassing the sublinear rates of prior theory.

Confident conflicts in LLM distillation are mitigated by dynamically mixing student and teacher distributions based on student entropy, boosting reasoning and preserving general capabilities.

Decoupling update magnitude from gradient norm in SGD initially accelerates learning but leads to marginal stability and distinct scaling regimes determined by model width and batch size.

Tabular ICL models can achieve near full-context performance with drastically reduced inference costs by aligning intermediate activations, even with minimal training data and no GPU.

Local AI models now predict user well-being from passive computer activity, offering explainable feedback directly within a workplace sensing application.

Binary sources allow for strictly smaller optimal representations in the Information Bottleneck than previously thought.

A single token can bypass most safety filters, revealing a critical vulnerability in current LLM alignment techniques.

LLM inference cost plummets as a novel output-aware KV cache retrieval method bypasses attention score approximation errors, achieving competitive performance with minimal overhead.

Hyper-connections create a fading memory in Transformers, with the effective context horizon directly tied to the second singular value of the stream-mixing matrices.

Generative auctions can now achieve externality-aware whole-feed optimization with exact strategyproofness using a transformer, overcoming a critical deployment barrier.

Single-neuron models can now capture multimodal voltage responses, including coexisting firing/non-firing states and depolarization block modes, a feat previously requiring massive ensembles or failing with neural operators.

Deterministic weather foundation models can be efficiently transformed into powerful probabilistic forecasters, generating skillful 15-day forecasts in minutes.

Nighttime anomaly detection is significantly enhanced by fusing M-band and DNB satellite signals, achieving higher accuracy and broader applicability than thermal-only methods.

Clinical ML models are only as good as their filtered data, yet current practices obscure crucial scientific decisions that shape task complexity and performance.

Amortized spike-and-slab inference for linear models with interactions achieves state-of-the-art variable selection accuracy with a single forward pass, eliminating costly per-dataset MCMC.

LLM sycophancy is not an intrinsic property but a malleable behavior shaped by subtle variations

Network traffic alone can reveal LLM agent risks and granular behaviors, but current analysis methods fail to distinguish subtle malicious activities or semantically similar scenarios.

Freezing the input factor A in LoRA, especially on top singular directions, unlocks substantial performance gains over freezing the output factor B, even at rank 2.

Frequency-specific conditional relationships in multivariate time series are unlocked by using inverse cross-spectral density matrices as graph shift operators.

Asynchronous execution fundamentally alters action semantics for world-model controllers, causing planning models to hallucinate future actions and recurrent models to misattribute transitions.

Explicitly modeling the temporal rate of change in spatio-temporal graph models unlocks superior prediction accuracy and pattern preservation for propagating phenomena.

The optimal LoRA rank is not intrinsic to the model but dynamically dictated by the data's geometric properties, enabling principled data selection for efficient fine-tuning.

A single space token transforms Transformers from arithmetic failures to 89% accurate multipliers by enabling learnable carry propagation.

Autoregressive latent priors learned directly from data, not fixed Gaussians, unlock VAEs for high-fidelity sequential generation with diffusion-level quality and faster inference.

Transformer attention applied to geometry-aware segment tokens unlocks 2459x speedup for electromigration stress prediction with sub-8% error.

Multi-omics cancer analysis becomes practical even with missing data, achieving state-of-the-art reconstruction and classification via a novel representation learning approach.

Hierarchical procedural understanding unlocks a 9.6%p performance boost in proactive AI assistance, enabling adaptive guidance that outperforms in-context learning by 57.1%p.

TSFMs for predictive control fail to capture system dynamics without explicit control excitation in their inference context, even with low forecasting error.

Pretrained time-series foundation models offer surprisingly inconsistent benefits for EMG tasks, with full fine-tuning often outperforming linear probing and few-shot adaptation only partially recovering performance on new users.

Agentic loops can now reliably stop with provable guarantees on proposal quality, even with imperfect verifiers.

Recover a full Pareto stationary front of multi-objective solutions with the computational cost of standard single-objective training.

Text generation evaluation can now directly measure distributional divergence using hidden states and double sliced Wasserstein distance, bypassing perplexity's limitations for non-autoregressive models.

Legal LLMs trained on surface features like citation count and legalese density learn to strategically withhold answers, becoming maximally "lawyerly" but minimally useful.

Equivariant potentials can achieve transformer-level accuracy with 35x fewer parameters and 10x faster inference by leveraging Cartesian tensor products.

Physics-based secure key distribution can now sustainably support federated learning for medical imaging, reducing key material needs by 35x without sacrificing accuracy.

The pursuit of LLM performance is shifting from raw scale to strategic resource allocation, revealing a critical need to re-evaluate scaling laws for parameter-efficient, compute-optimal training in constrained environments.

FairProp embeds fairness directly into GNN message passing, achieving state-of-the-art fairness-utility trade-offs by jointly optimizing group means and within-group covariance.

Metaheuristic search for Fourier fine-tuning coordinates unlocks substantial parameter efficiency gains, outperforming LoRA and fixed sampling strategies.

Multi-step reasoning errors amplify exponentially with step count, but transformers can be trained to achieve stable, near-zero inference loss over arbitrarily long chains.

Prompt injection defenses can degrade LLM agent utility by altering benign behavior; RAISED achieves robustness without this cost through self-distillation on generated scenarios.

Most reported gains in crop disease classification from environmental sensors are illusory, driven by simple session memorization rather than genuine multimodal learning.

Kernel ensembles with Survival Forests and Beran estimators significantly outperform single models for patient-specific survival prediction.

Deep GNNs can achieve superior depth scaling and accuracy by leveraging controlled Turing instability or stable near-critical propagation to combat over-smoothing.

A novel bandit algorithm achieves near-optimal dynamic minimax regret for LLM post-training, improving worst-group robustness without re-evaluating historical data or altering standard training pipelines.

Faithful concept bottleneck explanations are impossible without including all concepts, but probabilistic modeling and group lasso enable compact, faithful explanations.

Medical LLMs can achieve superior QA performance by dynamically selecting specialized adaptation components based on question context and clinical tags, surpassing static parameter-efficient methods.

Merging transformers sequentially along their execution path, rather than layer-by-layer, dramatically reduces error accumulation and unlocks principled multi-task capacity expansion.

Influence functions, not just activation averages, unlock more thematic and robust concept steering in LLMs.

Personalized suicide risk forecasting is significantly improved by embedding patients in a latent space that models similarity, outperforming pooled and per-patient models by up to 30% for first-occurrence event prediction.

Auditable text watermarking is now feasible, offering a practical compliance path for generative AI by quantifying detection uncertainty rather than claiming impossible robustness.

Deterministic midpoint methods for Langevin sampling can achieve superior sample complexity in high dimensions compared to their randomized counterparts.

Most LLM vulnerability reports are right about the bug but wrong about *why*, with free-form reasoning hiding critical flaws that VERA's structured audit uncovers.

Unreported sinkhole locations are effectively treated as right-censored, unlocking continuous, time-aware susceptibility predictions from multimodal data.

A dual VAE architecture effectively bridges the sim-to-real gap for robotic navigation, enabling high performance with minimal real-world data.

Wi-Fi sensing accuracy jumps to 92.83% by decomposing CSI into trajectory-guided tokens that capture complex temporal dynamics.

Spurious features are not just learned, they actively undermine worst-group accuracy during head retraining, even with existing robust methods.