✗ Failed on benchmark
2026
Replace a slow sequence of resolvent or contractive fixed-point updates by a blockwise averaged-reflection extrapolation. The method computes reflected iterates R^j y_0, averages them with equal weights, and uses the result as the next macro-iterate. Unlike unconstrained Anderson acceleration, this construction has a uniform residual guarantee for every maximal monotone operator.
Useful8/10
Difficulty4/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace an unconstrained neural decision head followed by clipping, projection, or penalty losses with a Legendre-regularized optimizer layer that always returns a feasible decision. The layer maps network-produced latent scores smoothly into the relative interior of a convex feasible region and can be differentiated using the inverse Hessian of the regularizer.
Useful8/10
Difficulty5/10
Novelty6/10
✓✓ Beats tuned baseline
2026
Construct a prescribed-performance funnel directly from state-only demonstrations, then train a state-feedback neural network whose output is bounded and whose gain is optimized to keep the tracking error inside that funnel. The controller should not imitate actions; it should reproduce the demonstrated transient and steady-state error geometry while explicitly reducing feedback authority whenever actuator saturation would make the funnel infeasible.
Useful8/10
Difficulty6/10
Novelty7/10
△ Mechanism confirmed, baseline not beaten
2026
Wrap a neural policy with a safety filter that minimally modifies its action so that a control-barrier inequality remains satisfied under bounded model mismatch and actuator saturation. Estimate mismatch between a learned plant or reference model and observed transitions online, then enlarge a conservative error margin and shrink the admissible safe set before solving the filter. The neural policy is unchanged when its action is safe, but receives a principled correction near state or action…
Useful8/10
Difficulty5/10
Novelty6/10
✗ Failed on benchmark
2026
When candidate rewards are noisy, do not score every candidate once and immediately retrain on the apparent elites. Sequentially resample only candidates near the selection boundary and certify their ordering with a mean, median-of-means, trimmed mean, or sign test selected according to the observed tail behavior.
Useful8/10
Difficulty4/10
Novelty5/10
✗ Failed on benchmark
2026
Equip a learned dynamics model with an adaptive parameter estimate and an explicit component-wise uncertainty box. Require a nominal backup-policy rollout to remain inside a safety margin equal to the rollout's worst-case parameter sensitivity, producing a conservative filter for reinforcement learning and world-model planning that becomes less conservative as the model identifies its parameters.
Useful8/10
Difficulty6/10
Novelty7/10
△ Mechanism confirmed, baseline not beaten
2026
Replace the ordinary average of bootstrapped Q-learning targets by a median-of-means estimator. For each current state-action anchor, divide repeated transition samples into blocks, average the target within each block, and take the median of the block averages; a minority of arbitrarily corrupted reward or next-state observations then affects fewer than half of the block estimates. For neural Q-learning, the same construction can be applied either to repeated samples for identical or nearby…
Useful8/10
Difficulty4/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Use fast memory as read-only scratch state during the internal pondering iterations of a recurrent block, and apply memory writes only after the latent computation has halted or crossed a write gate. This prevents the transition operator from changing while it is being iterated, reducing self-corruption of the evidence used for subsequent reasoning.
Useful8/10
Difficulty4/10
Novelty6/10
✗ Mechanism failed
2026
Attach a hard control-barrier-function quadratic-program safety filter to a neural policy, but solve the filter with operator splitting and differentiate through its fixed-point map using projection Jacobian-vector products. The network learns the nominal action and task objective end to end, while the deployed action remains the feasible filtered action rather than an unconstrained penalty-based approximation.
Useful8/10
Difficulty6/10
Novelty6/10
✗ Failed on benchmark
2026
Replace an unconstrained recurrent or neural-ODE state update with a copy of the known or learned plant dynamics plus a neural output-error correction, and train both the correction and a contraction metric using a pointwise matrix inequality penalty. The resulting observer should forget initialization exponentially and should amplify measurement noise by a quantitatively bounded factor rather than exhibiting unconstrained recurrent error growth.
Useful8/10
Difficulty6/10
Novelty7/10
✗ Mechanism failed
2026
Replace ordinary empirical-risk minimization on pooled heterogeneous data with worst-case conditional risk over joint distributions that remain close to every source under an optimal-transport budget. The adversary transports source context-label pairs toward high-loss, target-event-like examples, while source-specific radii prevent arbitrary shifts. This should improve performance on rare target contexts and unseen domains without requiring abundant target labels.
Useful8/10
Difficulty6/10
Novelty5/10
✗ Failed on benchmark
2026
Augment a neural state-space model or neural ODE with a low-dimensional control residual that keeps its hidden state inside a sequence of time-varying convex sets encoding temporal requirements. At each integration step, solve a small quadratic program that minimally changes the network dynamics while enforcing an inward-pointing condition on every active convex-set face, producing robustly constrained long-horizon rollouts.
Useful8/10
Difficulty6/10
Novelty7/10
△ Mechanism confirmed, baseline not beaten
2026
Before quantizing a matrix product, reparameterize its factors as A'=AT and B'=T^{-1}B, preserving the exact full-precision product while changing the quantization difficulty of each factor. Choose a positive diagonal T=diag(t_1,...,t_K) that minimizes predicted post-quantization product error, rather than using output-channel scaling or a fixed heuristic grid. The gauge can be shared across several products when transformed-copy cost matters.
Useful8/10
Difficulty5/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace a fixed optimizer preconditioner with a diagonal matrix selected by an online convex optimizer. A gradient predictor supplies the direction, while a linear-loss regret update learns coordinate-wise gains that favor transformations aligned with the realized stochastic gradient. The method retains the identity preconditioner as an explicit comparator, so it can be tested for negative regret and improvement over ordinary SGD.
Useful8/10
Difficulty4/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Implement causal linear attention in chunks and combine chunk summaries with an associative scan instead of carrying the recurrent state through all chunks sequentially. This preserves the exact causal computation while reducing inter-chunk dependency depth from the number of chunks to its logarithm, enabling substantially more GPU parallelism for long-context training and prefill.
Useful8/10
Difficulty6/10
Novelty6/10
✗ Failed on benchmark
2026
Insert an online errors-in-variables subspace estimator into a latent state-space neural network. A fixed recent window of encoder features and controls is used to estimate a noise-corrected low-dimensional state subspace and refit the latent transition and readout matrices, allowing the model to follow sensor degradation or changing operating conditions without replaying the entire dataset.
Useful8/10
Difficulty6/10
Novelty6/10
✗ Failed on benchmark
2026
Model a multi-timescale optimizer as a controlled dynamical system and use several Lyapunov-like quantities to regulate loss, momentum energy, and constraint violation simultaneously. The explicit high-order control-Lyapunov feedback becomes a low-cost correction to an SGD-momentum or Adam step. A Hurwitz comparison matrix supplies a measurable stability certificate and predicts the decay rate of the controlled training dynamics.
Useful8/10
Difficulty6/10
Novelty8/10
✗ Mechanism failed
2026
Replace an unconstrained recurrent update or neural-ODE vector field with a nominal learned control plus an explicit high-order barrier correction. The correction enforces hidden-state safety even when the control affects the safety variable only after several time derivatives. A quadratic-program projection preserves the nominal network output whenever the learned dynamics already satisfy the barrier inequality.
Useful8/10
Difficulty6/10
Novelty7/10
✗ Failed on benchmark
2026
Treat neural modules as interconnected dynamical subsystems and estimate the gain from every module input to every neighboring module output. Replace an expensive global Jacobian spectral-radius calculation by decentralized directed-cycle tests inside clusters and path-gain tests between clusters. Penalizing violations during training should prevent exploding recurrent trajectories while retaining less conservative behavior than constraining every individual block independently.
Useful8/10
Difficulty6/10
Novelty6/10
✗ Failed on benchmark
2026
Wrap a neural policy with a control-barrier safety layer whose constraints use an online estimate of model mismatch or environmental disturbance. Instead of enforcing a fixed worst-case bound at every state, the layer reconstructs the current effective dynamics from an extended state observer and adds only the margin required by the remaining estimation error.
Useful8/10
Difficulty6/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Use the Gaussian trajectory predictor inside an inference-time planner or model-based reinforcement-learning policy, optimizing a nominal action sequence together with affine feedback gains against predicted disturbances. The resulting controller reacts to realized model residuals rather than relying on open-loop neural rollouts, while preserving a convex quadratic structure when the prediction map and covariance are frozen.
Useful8/10
Difficulty6/10
Novelty5/10
✗ Mechanism failed
2026
Train a primal state network and a dual flux network jointly, using the convex primal-dual gap as the main loss and as an a posteriori certificate of state error. Unlike a strong residual, the certificate is based on monotonicity and convex duality, so it can remain informative even when differentiating rapidly oscillatory coefficients would amplify noise by $1/\varepsilon$.
Useful8/10
Difficulty6/10
Novelty7/10
✗ Failed on benchmark
2026
Replace a single value critic with graph-indexed critics V_alpha and graph-indexed policy heads pi_A, where the labeled graph covers every possible environment mode at every step. Train sampled Bellman inequalities rather than only equality-based temporal-difference errors, and select the policy head using the paper's min-max reachability rule. This targets robust RL settings in which the transition mode can change arbitrarily, such as unknown actuator regimes, domain randomization modes, or…
Useful8/10
Difficulty5/10
Novelty7/10
△ Mechanism confirmed, baseline not beaten
2026
Replace layer-local reconstruction in post-training quantization with a sequential objective that explicitly cancels the error already accumulated by the quantized prefix. For each layer, quantize its weights so that its local residual approximately negates the propagated incoming deviation, preserving the teacher trajectory even when the codebook is binary or 4-bit.
Useful8/10
Difficulty5/10
Novelty7/10