Daily report — 2026-10-07

Off-policy evaluation (OPE) for contextual bandit policies becomes challenging when action-level importance weighting incurs excessive variance. Doubly robust (DR) estimation remains unbiased under common support but retains these high-variance action-level weights.

Claimed: none extracted

Evidence: + cheap/closed-form ("closed form"); + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + public benchmarks ("benchmarks"); + concrete method ("closed form"); + formal statement ("theorem"); no public code found; + keywords: bandit

heuristic

Combinatorial Thompson sampling (CTS) draws independent posterior samples for every arm, so its exploration dynamics ignore any relation among arms. We study a minimal change to those dynamics: an LLM is queried once for a partition of the arms, the partition becomes a positive-definite correlation matrix $Σ$ through an RBF kernel on cluster ranks, and the per-round posterior sample is drawn with

Claimed: none extracted

Evidence: - LLM-based ("LLM"); + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + public benchmarks ("benchmark"); + explicit algorithm/pseudocode ("pseudocode"); + formal statement ("bound"); + claims simplicity ("simple"); no public code found; + keywords: sampling, kernel, bandit

heuristic

We study linear bandits under exact sliding-window constraints, where every consecutive block of actions must belong to a prescribed feasible set. In the offline setting, where the reward function is known, we show that convexity and cyclic-shift invariance make a stationary solution optimal when $w\mid T$ and within an additive $O(w)$ gap otherwise.

Claimed: none extracted

Evidence: + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + public benchmarks ("benchmarks"); + explicit algorithm/pseudocode ("Algorithm 1"); + formal statement ("bound"); no public code found; + keywords: bandit

heuristic

CNet is a C++/CUDA framework for building and training deep complex-valued neural networks (CVNNs) and, more generally, for optimizing complex-valued functions by gradient descent with Wirtinger (CR-calculus) derivatives. It takes a physics-native stance: a network is a cascade of complex -- and often unitary (the DFT) -- operations acting on an amplitude vector, and classification is a Born-rule

Claimed: none extracted

Evidence: + CPU-friendly ("CPU"); + concrete method ("closed form"); - system/framework (many moving parts) ("Framework"); + code: https://github.com/crasmarum/CNet; + keywords: optimizer, kernel

heuristic

Concept drift threatens production machine learning, yet the empirical behavior of multivariate two-sample drift detectors at scale remains under-characterized. Existing benchmarks rarely address the hundreds of millions of rows and high-cardinality features typical of industrial-operational datasets.

Claimed: none extracted

Evidence: + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + public benchmarks ("benchmarks"); no public code found; + keywords: sampling, kernel, ranking

heuristic

In cross-lingual zero-shot text-to-speech, the accent of the reference leaks into the target speech. We propose accent analogy guidance (AAG), a training-free sampler term that subtracts an accent direction estimated from the model's own predictions for one synthetic voice rendered in both languages, so the voice cancels and only the accent remains.

Claimed: none extracted

Evidence: - LLM-based ("LLM"); + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + code: https://github.com/yoomee-cho/accent-analogy-guidance; + 4 HF upvotes

heuristic

Contextual priors improve neural-to-language decoding by reshaping candidate scores. However, confidence is read from the same reshaped scores, so the errors a prior leaves behind can become more confident with no change in accuracy to reveal it.

Claimed: none extracted

Evidence: + formal statement ("bound"); + code: https://github.com/AmadeusFake/NeuDecodingConfReversal; + keywords: retrieval, ranking

heuristic

Retrieval-augmented generation systems can route difficult queries to deeper context, but batch deployments must allocate a shared token budget across calls whose costs vary by query. We formulate selective escalation as finite-batch allocation for financial document question answering.

Claimed: none extracted

Evidence: + formal statement ("bound"); + code: https://github.com/junru-zhu/token-budgeted-escalation-financial-qa; + keywords: retrieval, ranking

heuristic

Motivated by sequence-to-sequence transport in the context time-series domain adaptation, we study the problem of transportation between trajectories of Markov processes. Given a limited number of trajectories from source distribution and the target distribution, we formulate a flow matching based algorithm which learns a transport map from the source to target trajectory distribution, while prese

Claimed: accuracy 43.87%

Evidence: + toy/synthetic experiments ("synthetic"); + small/synthetic data ("synthetic"); + formal statement ("bound"); no public code found; + keywords: retrieval

heuristic

Tool-using agents make consequential changes to external state, yet correct outcomes do not guarantee that their actions were supported by evidence established beforehand. We study where this evidence-to-action chain breaks as agents move from deciding whether to act to executing single actions and dependent workflows.

Claimed: none extracted

Evidence: + formal statement ("bound"); + code: https://github.com/caoshidong66/safeact; + 30 HF upvotes

heuristic