arXiv subjects

Rong Jiang

Publications and source records attributed to Rong Jiang.

9 recordsLinked to original sources

arXiv2026

STEC: Evidence Compression for Deep Search in Open-domain Multi-Hop QA

In open-domain multi-hop question answering (QA), LLM-based search agents offer a promising approach to knowledge-intensive QA by combining retrieval with reasoning. Existing methods mainly improve open-domain multi-hop QA through reasoning paradigms, retrieval interaction, and search strategy optimization. However, using multiple search trajectories introduces a challenging final answer selection problem. Different trajectories may support different candidates, and the retrieved information can be heterogeneous, redundant, incomplete, or conflicting. Directly comparing raw trajectories exposes the verifier to noisy and unaligned content, while comparing answer strings ignores the evidence supporting each candidate, making reliable final selection difficult. To address this challenge, we propose STEC, an evidence compression framework for final answer selection in multi-hop QA. STEC selects the final answer from the existing candidate set through two mechanisms: (1) Answer-Level Evidence Compression, which groups trajectories by normalized answer identity and converts each answer group into a candidate-specific evidence representation; and (2) Evidence-Guided Answer Verification, which compares these representations and selects the final answer from the candidate set. The design shifts final selection from raw trajectory comparison to candidate-level evidence comparison. We evaluate STEC on four open-domain multi-hop QA benchmarks against representative baselines. Experimental results show that STEC performs best overall among the compared methods, and ablation results provide evidence that answer-level evidence compression contributes to final answer selection.

cs.AI↗

arXiv2025

The Adaptivity Barrier in Batched Nonparametric Bandits: Sharp Characterization of the Price of Unknown Margin

We study batched nonparametric contextual bandits under a margin condition when the margin parameter $\alpha$ is unknown. To capture the statistical cost of this ignorance, we introduce the regret inflation criterion, defined as the ratio between the regret of an adaptive algorithm and that of an oracle knowing $\alpha$. We show that the optimal regret inflation grows polynomially with the horizon $T$, with exponent given by the value of a convex optimization problem that depends on the dimension, smoothness, and number of batches $M$. Moreover, the minimizer of this optimization problem directly prescribes the batch allocation and exploration strategy of a rate-optimal algorithm. Building on this principle, we develop RoBIN (RObust batched algorithm with adaptive BINning), which achieves the optimal regret inflation up to polylogarithmic factors. These results reveal a new adaptivity barrier: under batching, adaptation to an unknown margin parameter inevitably incurs a polynomial penalty, sharply characterized by a variational problem. Remarkably, this barrier vanishes once the number of batches exceeds order $\log \log T$; with only a doubly logarithmic number of updates, one can recover the oracle regret rate up to polylogarithmic factors.

math.ST↗

arXiv2025

Exemplar-Guided Planing: Enhanced LLM Agent for KGQA

Large Language Models (LLMs) as interactive agents show significant promise in Knowledge Graph Question Answering (KGQA) but often struggle with the semantic gap between natural language queries and structured knowledge graph (KG) representations. This leads to suboptimal planning and inefficient exploration on KG, while training-free approaches often underutilize valuable reasoning patterns in training data. To address these limitations, we propose a novel framework, Exemplar-Guided Planning (EGP), which enhances the planning capabilities of LLM agents for KGQA. EGP first preprocesses the training set questions via entity templating to normalize semantic variations. It then retrieves highly similar exemplary questions and their successful reasoning paths from this preprocessed set using semantic embeddings and an efficient FAISS index. These retrieved exemplars dynamically guide the LLM's planning process in two key phases: (1) Task Decomposition, by aligning generated sub-objectives with proven reasoning steps, and (2) Relation Exploration, by providing high-quality auxiliary information to improve relation pruning accuracy. Additionally, we introduce a Smart Lookahead mechanism during relation exploration to improve efficiency by preemptively exploring promising paths and potentially terminating exploration earlier. We apply EGP to the Plan-on-Graph (PoG) framework, termed PoG-EGP. Extensive experiments on two real-world KGQA datasets, WebQSP and CWQ, demonstrate that PoG-EGP significantly improves over the baseline PoG system and other compared methods.

cs.CL↗

arXiv2025

Semi-supervised learning for linear extremile regression

Extremile regression, as a least squares analog of quantile regression, is potentially useful tool for modeling and understanding the extreme tails of a distribution. However, existing extremile regression methods, as nonparametric approaches, may face challenges in high-dimensional settings due to data sparsity, computational inefficiency, and the risk of overfitting. While linear regression serves as the foundation for many other statistical and machine learning models due to its simplicity, interpretability, and relatively easy implementation, particularly in high-dimensional settings, this paper introduces a novel definition of linear extremile regression along with an accompanying estimation methodology. The regression coefficient estimators of this method achieve $\sqrt{n}$-consistency, which nonparametric extremile regression may not provide. In particular, while semi-supervised learning can leverage unlabeled data to make more accurate predictions and avoid overfitting to small labeled datasets in high-dimensional spaces, we propose a semi-supervised learning approach to enhance estimation efficiency, even when the specified linear extremile regression model may be misspecified. Both simulation studies and real data analyses demonstrate the finite-sample performance of our proposed methods.

stat.ME↗

arXiv2025

Average quantile regression: a new non-mean regression model and coherent risk measure

Regression models that go beyond the mean, alongside coherent risk measures, have been important tools in modern data analysis. This paper introduces the innovative concept of Average Quantile Regression (AQR), which is smooth at the quantile-like level, comonotonically additive, and explicitly accounts for the severity of tail losses relative to quantile regression. AQR serves as a versatile regression model capable of describing distributional information across all positions, akin to quantile regression, yet offering enhanced interpretability compared to expectiles. Numerous traditional regression models and coherent risk measures can be regarded as special cases of AQR. As a flexible non-parametric regression model, AQR demonstrates outstanding performance in analyzing high-dimensional and large datasets, particularly those generated by distributed systems, and provides a convenient framework for their statistical analysis. The corresponding estimators are rigorously derived, and their asymptotic properties are thoroughly developed. In a risk management context, the case study confirms AQR's effectiveness in risk assessment and portfolio optimization.

math.ST↗

arXiv2024

Batched Nonparametric Contextual Bandits

We study nonparametric contextual bandits under batch constraints, where the expected reward for each action is modeled as a smooth function of covariates, and the policy updates are made at the end of each batch of observations. We establish a minimax regret lower bound for this setting and propose a novel batch learning algorithm that achieves the optimal regret (up to logarithmic factors). In essence, our procedure dynamically splits the covariate space into smaller bins, carefully aligning their widths with the batch size. Our theoretical results suggest that for nonparametric contextual bandits, a nearly constant number of policy updates can attain optimal regret in the fully online setting.

math.ST↗

arXiv2023

Root n consistent extremile regression and its supervised and semi-supervised learning

Extremile (Daouia, Gijbels and Stupfler,2019) is a novel and coherent measure of risk, determined by weighted expectations rather than tail probabilities. It finds application in risk management, and, in contrast to quantiles, it fulfills the axioms of consistency, taking into account the severity of tail losses. However, existing studies (Daouia, Gijbels and Stupfler,2019,2022) on extremile involve unknown distribution functions, making it challenging to obtain a root n-consistent estimator for unknown parameters in linear extremile regression. This article introduces a new definition of linear extremile regression and its estimation method, where the estimator is root n-consistent. Additionally, while the analysis of unlabeled data for extremes presents a significant challenge and is currently a topic of great interest in machine learning for various classification problems, we have developed a semi-supervised framework for the proposed extremile regression using unlabeled data. This framework can also enhance estimation accuracy under model misspecification. Both simulations and real data analyses have been conducted to illustrate the finite sample performance of the proposed methods.

stat.ME↗

arXiv2023

Modile as a conservative tail risk measurer: the solution of an optimisation problem with 0-1 loss function

Quantiles and expectiles, which are two important concepts and tools in tail risk measurements, can be regarded as an extension of median and mean, respectively. Both of these tail risk measurers can actually be embedded in a common framework of $L_p$ optimization with the absolute loss function ($p=1$) and quadratic loss function ($p=2$), respectively. When 0-1 loss function is frequently used in statistics, machine learning and decision theory, this paper introduces an 0-1 loss function based $L_0$ optimisation problem for tail risk measure and names its solution as modile, which can be regarded as an extension of mode. Mode, as another measure of central tendency, is more robust than expectiles with outliers and easy to compute than quantiles. However, mode based extension for tail risk measure is new. This paper shows that the proposed modiles are not only more conservative than quantiles and expectiles for skewed and heavy-tailed distributions, but also providing or including the unique interpretation of these measures. Further, the modiles can be regarded as a type of generalized quantiles and doubly truncated tail measure whcih have recently attracted a lot of attention in the literature. The asymptotic properties of the corresponding sample-based estimators of modiles are provided, which, together with numerical analysis results, show that the proposed modiles are promising for tail measurement.

math.ST↗

arXiv2015

Negative Group Velocity Pulse Propagation Through a Left-Handed Transmission Line

In this paper, the microwave pulse propagation transferred through a left-handed transmission line using Complementary Omega-Like Structures (COLS) loaded was studied. There was a stop band in transmission from 5.6GHz to 6.1GHz, and the anomalous dispersion was causes in this band. Negative group velocity corresponds to the case in which the peak of the pulse exited before the peak of the incident pulse had entered the sample. The negative group velocity reached (-0.27c~-1.85c).

physics.optics↗