Daily report — 2026-09-25
146 new papers screened · 20 analyzed · 0 reproduction runs · markdown
Reproductions
| status | paper | backend | claims matched |
|---|---|---|---|
| No reproduction runs today (opt-in). | |||
Top candidates
| score | paper | feasibility |
|---|---|---|
| 67.9 | Global Convergence of Third-Order Langevin Dynamics for Non-Convex Optimization via Simulated Annealing We study global convergence guarantees of third-order Langevin dynamics for non-convex optimization via simulated annealing with fixed friction and decreasing noise. An explicit three-block distorted entropy transfers dissipation from the noisy auxiliary variable to the full state. Claimed: none extracted Evidence: + cheap/closed-form ("Convex"); + toy/synthetic experiments ("Simulated"); + small/synthetic data ("Simulated"); + formal statement ("Convergence"); + claims simplicity ("simple"); + code: https://github.com/gagawjbytw/simulated-annealing-third-order-langevin | 8/10 heuristic |
| 56.8 | Orbital Error Dynamics: Self-Organized Criticality, Ephemeral Parameter Resonance, and Non-Linear Biological Ontologies in Zero-Storage Neural Synthesis Modern deep neural networks treat parameters as static floating-point matrices stored in physical memory, incurring Von Neumann memory bottlenecks and representation collapse. We formulate Orbital Error Dynamics (OED), an analytical framework wherein synaptic weights are not stored masses (O(W)), but transient topological resonances (O(1)) derived procedurally from the complex quadratic polynomial Claimed: accuracy 8.0 Evidence: + cheap/closed-form ("convex"); + explicit algorithm/pseudocode ("Algorithm 1"); + formal statement ("Convergence"); - system/framework (many moving parts) ("framework"); + code: https://github.com/pCwOrM/mandelbrot-fractal-neural-synthesis | 6/10 heuristic |
| 56.3 | OBLIQ-IR: Training a Dense Retriever for Oblique Queries Oblique retrieval, as exemplified by OBLIQ-Bench, asks a retriever to find documents whose relevance is determined by a latent attribute (an implicit stance, an analogous reasoning technique, an authorial fingerprint, or a vague tip-of-the-tongue recollection) that has little or no surface expression in the document. State-of-the-art dense encoders and agentic search pipelines built around frontie Claimed: NDCG@10 0.211; NDCG@10 0.171; NDCG@10 2.0; NDCG@10 0.222; NDCG@1 0.0 Evidence: + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); - system/framework (many moving parts) ("agentic"); + code: https://github.com/DataScienceUIBK/obliq-ir; + keywords: retrieval | 6/10 heuristic |
| 54.2 | MF-SCBO : Multi-fidelity Scalable Constrained Bayesian Optimization Many real-world optimization problems rely on expensive simulations or experiments, making the efficient use of available data essential. Multi-fidelity optimization of high-dimensional black-box functions subject to black-box constraints is increasingly relevant as the cost of objective evaluations continues to rise in applications such as machine learning, engineering, and control. Claimed: none extracted Evidence: + toy/synthetic experiments ("simulations"); + public benchmarks ("benchmark"); + explicit algorithm/pseudocode ("Algorithm 1"); + formal statement ("convergence"); + claims simplicity ("simple"); no public code found; + keywords: sampling | 7/10 heuristic |
| 53.7 | Residual Correlation as a Diagnostic for Joint-Uncertainty Gains from GP Coregionalisation In multi-target regression, correlated targets are often coupled through multi-output Gaussian processes with an intrinsic model of coregionalisation (GP-ICM), assuming that sharing statistical strength improves overall performance. In practice, the benefits are inconsistent. Claimed: none extracted Evidence: + toy/synthetic experiments ("synthetic"); + CPU-friendly ("lightweight"); + small/synthetic data ("synthetic"); + public benchmarks ("benchmarks"); + formal statement ("bound"); no public code found | 7/10 heuristic |
| 52.6 | Exact Bayes Regret and Asymptotic Optimality in High-Dimensional Gaussian Bandits We study Bayesian linear bandits with an isotropic Gaussian parameter, independent Gaussian candidate arms, and Gaussian reward noise when the horizon is proportional to the dimension. The normalized posterior uncertainty has an explicit limit that is uniform over all causal policies. Claimed: none extracted Evidence: + cheap/closed-form ("Closed-form"); + concrete method ("Closed-form"); + formal statement ("bound"); no public code found; + keywords: sampling, bandit | 6/10 heuristic |
| 52.6 | Temporal Gradient Inversion for Private Trajectory Reconstruction in Embodied Reinforcement Learning Distributed learning in embodied reinforcement-learning agents offers a degree of privacy by retaining raw sensor data on-device and transmitting only policy gradients to the server. Yet temporal structure can amplify this leakage beyond single-frame attacks. Claimed: PSNR 3.0 Evidence: + cheap/closed-form ("closed-form"); + concrete method ("closed-form"); + formal statement ("bound"); + claims simplicity ("simple"); no public code found; + keywords: regularization | 7/10 heuristic |
| 52.6 | Direct Message Approximation (DMA): A Consistency-Based Framework for Tractable Approximate Inference on Factor Graphs Approximate message passing on factor graphs underlies two dominant families of probabilistic inference algorithms: expectation propagation (EP) and variational message passing (VMP). Both methods approximate the marginal at each factor edge, forcing an iterative round-robin schedule, risking negative-precision messages, and, for VMP, collapsing to point estimates at Dirac-delta factors. Claimed: none extracted Evidence: + cheap/closed-form ("closed-form"); + explicit algorithm/pseudocode ("Algorithm 1"); + concrete method ("closed-form"); + formal statement ("We prove"); - system/framework (many moving parts) ("Framework"); no public code found; + keywords: variational | 7/10 heuristic |
| 49.5 | SmallReason-ColBERT: An Ultra-Small Late-Interaction Retriever for Reasoning Intensive Retrieval Reasoning-intensive retrieval remains difficult for small models. Compact public ColBERTs are usually trained on general-purpose corpora and underperform reasoning-tuned 150M+ baselines on BRIGHT~\cite{bright} by several nDCG@10 points. Claimed: nDCG@1 0.0; nDCG@10 1.21 Evidence: + formal statement ("theorem"); + code: https://github.com/DataScienceUIBK/SmallReason-ColBERT; + keywords: retrieval | 5/10 heuristic |
| 46.8 | Decoupled Learning and Selection in Slate Recommendation for Privacy and Stability Under Noisy Scores We formalize slate recommendation as a randomized score learner followed by deterministic selection. First, an appropriately scoped differential-privacy guarantee passes through selection and its audit trace by post-processing. Claimed: none extracted Evidence: + toy/synthetic experiments ("simulations"); + formal statement ("bound"); + claims simplicity ("simple"); no public code found; + keywords: ranking | 6/10 heuristic |
| 46.3 | JevOut: Natural Context Can Flip Decision Models Dedicated decision models such as Jev map unstructured language to probability distributions over finite choices, allowing their outputs to directly route requests, select tools, and trigger actions. Yet real-world inputs rarely arrive in isolation: they come with background details and surrounding context. Claimed: none extracted Evidence: + code: https://github.com/xzx34/JevOut; + keywords: optimizer | 4/10 heuristic |
| 46.3 | The Fellowship of the Query: Learning Retrieval Actions Retrieval-augmented question answering requires control decisions about when to decompose a question, search, reformulate, extract evidence, synthesize facts, verify progress, and stop. We study whether trajectory fine-tuning can improve small language models (SLMs) as next-action controllers. Claimed: F1 0.6536; F1 0.7783 Evidence: + code: https://github.com/padas-lab-de/agent-action-controller; + keywords: retrieval | 4/10 heuristic |
| 45.8 | Ingest-Time Fact Compilation for Cost-Efficient and Reliable Question Answering over Revised Corpora Most agentic question answering (QA) systems do an important part of their semantic work at the worst possible time: every time someone asks a question. When a corpus contains revisions, drafts, revocations, deletions, and sources with different levels of authority, the model must reconstruct the governed current state on every read - then throw that work away and repeat it on the next query. Claimed: none extracted Evidence: + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + claims simplicity ("Simple"); - system/framework (many moving parts) ("agentic"); no public code found | 6/10 heuristic |
| 45.3 | A Nearly Quadratic Lower Bound for Linear Optimization over Convex Bodies in the Membership Oracle Model We prove nearly quadratic lower bounds for randomized algorithms for linear optimization and uniform sampling over convex bodies in the membership oracle model. For linear optimization, this matches the known nearly quadratic upper bound up to a polylog factor in the dimension. Claimed: none extracted Evidence: + cheap/closed-form ("Convex"); + formal statement ("Bound"); no public code found; + keywords: sampling | 6/10 heuristic |
| 44.7 | AERIAL: Adversarial Evaluation of Robustness in Accuracy-Preserving Low-Precision EEG Decoders Deployment-oriented compression is attractive for resource-constrained brain--computer interfaces (BCIs), but whether it changes adversarial vulnerability remains unclear. On BCI Competition IV-2a, we compare 32-bit floating-point (FP32) EEGNet and ShallowConvNet models with global magnitude pruning and simulated INT8 post training quantization (PTQ) and quantization-aware training (QAT) across ni Claimed: accuracy 22.0 Evidence: + toy/synthetic experiments ("simulated"); + small/synthetic data ("simulated"); no public code found | 6/10 heuristic |
| 44.7 | Claim-Gated Source-Risk Auditing for Generative Search A generative search answer can cite a supported passage yet omit a source relationship that changes its interpretation. We specify a claim-gated audit of the query-source-answer tuple. Claimed: none extracted Evidence: + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); no public code found | 6/10 heuristic |
| 40.0 | EvLink: Source-Grounded Evidence Linking for Graph RAG Graph-based Retrieval-Augmented Generation (GraphRAG) supports multi-hop reasoning by organizing corpora into structured graphs. However, graph reachability often captures semantic association rather than evidence support, so a reachable passage may still fail to justify a required cross-passage transition. Claimed: F1 2.4 Evidence: + public benchmarks ("benchmarks"); + claims simplicity ("simple"); no public code found; + keywords: retrieval | 5/10 heuristic |
| 38.9 | From Interests to Semantic IDs: Retrieval-Grounded Credit Assignment for Generative Recommendation Semantic IDs (SIDs) encode each catalog item as a short token sequence, enabling generative recommenders to predict the next item autoregressively. Reasoning-enhanced variants, an increasingly common extension, first generate a textual trace and then decode a next-item SID by beam search. Claimed: none extracted Evidence: no public code found; + keywords: retrieval, ranking | 4/10 heuristic |
| 37.9 | Learning Better Reasoning for Generative Recommendation with Semantic IDs Generative recommendation reformulates item retrieval as sequence generation, allowing a unified model to directly generate the next item from a user's interaction history. Semantic IDs further make this paradigm effective and scalable by representing each item as discrete codes, enabling knowledge sharing among semantically related items. Claimed: none extracted Evidence: + public benchmarks ("benchmarks"); - system/framework (many moving parts) ("framework"); no public code found; + keywords: retrieval, ranking | 4/10 heuristic |
| 36.8 | A Systematic Multi-Domain Evaluation of Document Retrievers Document retrieval is a crucial component of many modern AI systems, directly influencing their effectiveness, robustness, and fairness in downstream tasks. While recent years have seen a growing number of retrievers, comparative studies in the literature are typically limited in scope or focused on singular benchmarks, domains, or model families. Claimed: none extracted Evidence: + public benchmarks ("benchmarks"); no public code found; + keywords: retrieval | 4/10 heuristic |