Conditional Performance Guarantee for Large Reasoning Models

Large reasoning models have shown strong performance through extended chain-of-thought reasoning, yet their computational cost remains significant. Probably approximately correct (PAC) reasoning provides statistical guarantees for efficient reasoning by adaptively switching between thinking and non-thinking models, but the guarantee holds only in the marginal case and does not provide exact conditional coverage. We propose G-PAC reasoning, a practical framework that provides PAC-style guarantees at the group level by partitioning the input space. We develop two instantiations: Group PAC (G-PAC) reasoning for known group structures and Clustered PAC (C-PAC) reasoning for unknown groupings. We prove that both G-PAC and C-PAC achieve group-conditional risk control, and that grouping can strictly improve efficiency over marginal PAC reasoning in heterogeneous settings. Our experiments on diverse reasoning benchmarks demonstrate that G-PAC and C-PAC successfully achieve group-conditional risk control while maintaining substantial computational savings.

Anytime Safe PAC Efficient Reasoning

Large Reasoning Models (LRMs) have demonstrated remarkable performance on complex tasks but suffer from high computational costs and latency. While selective thinking strategies improve efficiency by routing easy queries to non-thinking models, existing approaches often incur uncontrollable errors, especially in online settings where the performance loss of a non-thinking model is only partially observed and data are non-stationary. To address this, we propose Betting Probably Approximately Correct (B-PAC) reasoning, a principled method that enables anytime safe and efficient online reasoning under partial feedback. Specifically, we utilize inverse propensity scoring estimators to construct test supermartingales for candidate thresholds, and then dynamically adjust the routing threshold based on the accumulated statistical evidence of safety. Theoretically, we establish the anytime-valid performance loss control and the efficiency of B-PAC reasoning. Extensive experiments demonstrate that B-PAC reasoning significantly reduces computational overhead, decreasing thinking model usage by up to 81.01\%, while controlling the performance loss below the user-specified level.

Natural Language-Driven Global Mapping of Martian Landforms

Planetary surfaces are typically analyzed using high-level semantic concepts in natural language, yet vast orbital image archives remain organized at the pixel level. This mismatch limits scalable, open-ended exploration of planetary surfaces. Here we present MarScope, a planetary-scale vision-language framework enabling natural language-driven, label-free mapping of Martian landforms. MarScope aligns planetary images and text in a shared semantic space, trained on over 200,000 curated image-text pairs. This framework transforms global geomorphic mapping on Mars by replacing pre-defined classifications with flexible semantic retrieval, enabling arbitrary user queries across the entire planet in 5 seconds with F1 scores up to 0.978. Applications further show that it extends beyond morphological classification to facilitate process-oriented analysis and similarity-based geomorphological mapping at a planetary scale. MarScope establishes a new paradigm where natural language serves as a direct interface for scientific discovery over massive geospatial datasets.

Multi-Condition Conformal Selection

Selecting high-quality candidates from large-scale datasets is critically important in resource-constrained applications such as drug discovery, precision medicine, and the alignment of large language models. While conformal selection methods offer a rigorous solution with False Discovery Rate (FDR) control, their applicability is confined to single-threshold scenarios (i.e., y > c) and overlooks practical needs for multi-condition selection, such as conjunctive or disjunctive conditions. In this work, we propose the Multi-Condition Conformal Selection (MCCS) algorithm, which extends conformal selection to scenarios with multiple conditions. In particular, we introduce a novel nonconformity score with regional monotonicity for conjunctive conditions and a global Benjamini-Hochberg (BH) procedure for disjunctive conditions, thereby establishing finite-sample FDR control with theoretical guarantees. The integration of these components enables the proposed method to achieve rigorous FDR-controlled selection in various multi-condition environments. Extensive experiments validate the superiority of MCCS over baselines, its generalizability across diverse condition combinations, different real-world modalities, and multi-task scalability.

Semi-Supervised Conformal Prediction With Unlabeled Nonconformity Score

Conformal prediction (CP) is a powerful framework for uncertainty quantification, providing prediction sets with coverage guarantees when calibrated on sufficient labeled data. However, in real-world applications where labeled data is often limited, standard CP can lead to coverage deviation and output overly large prediction sets. In this paper, we extend CP to the semi-supervised setting and propose SemiCP, leveraging both labeled data and unlabeled data for calibration. Specifically, we introduce a novel nonconformity score function, NNM, designed for unlabeled data. This function selects labeled data with similar pseudo-label scores to estimate nonconformity scores, integrating them into the calibration process to overcome sample size limitations. We theoretically demonstrate that, under mild assumptions, SemiCP provide asymptotically coverage guarantee for prediction sets. Extensive experiments further validate that our approach effectively reduces instability and inefficiency under limited calibration data, can be adapted to conditional coverage settings, and integrates seamlessly with existing CP methods.

Exploring Imbalanced Annotations for Effective In-Context Learning

Large language models (LLMs) have shown impressive performance on downstream tasks through in-context learning (ICL), which heavily relies on the demonstrations selected from annotated datasets. Existing selection methods may hinge on the distribution of annotated datasets, which can often be long-tailed in real-world scenarios. In this work, we show that imbalanced class distributions in annotated datasets significantly degrade the performance of ICL across various tasks and selection methods. Moreover, traditional rebalance methods fail to ameliorate the issue of class imbalance in ICL. Our method is motivated by decomposing the distributional differences between annotated and test datasets into two-component weights: class-wise weights and conditional bias. The key idea behind our method is to estimate the conditional bias by minimizing the empirical error on a balanced validation dataset and to employ the two-component weights to modify the original scoring functions during selection. Our approach can prevent selecting too many demonstrations from a single class while preserving the effectiveness of the original selection methods. Extensive experiments demonstrate the effectiveness of our method, improving the average accuracy by up to 5.46 on common benchmarks with imbalanced datasets.

Parametric Scaling Law of Tuning Bias in Conformal Prediction

Conformal prediction is a popular framework of uncertainty quantification that constructs prediction sets with coverage guarantees. To uphold the exchangeability assumption, many conformal prediction methods necessitate an additional holdout set for parameter tuning. Yet, the impact of violating this principle on coverage remains underexplored, making it ambiguous in practical applications. In this work, we empirically find that the tuning bias - the coverage gap introduced by leveraging the same dataset for tuning and calibration, is negligible for simple parameter tuning in many conformal prediction methods. In particular, we observe the scaling law of the tuning bias: this bias increases with parameter space complexity and decreases with calibration set size. Formally, we establish a theoretical framework to quantify the tuning bias and provide rigorous proof for the scaling law of the tuning bias by deriving its upper bound. In the end, we discuss how to reduce the tuning bias, guided by the theories we developed.

ChineseSafe: A Chinese Benchmark for Evaluating Safety in Large Language Models

With the rapid development of Large language models (LLMs), understanding the capabilities of LLMs in identifying unsafe content has become increasingly important. While previous works have introduced several benchmarks to evaluate the safety risk of LLMs, the community still has a limited understanding of current LLMs' capability to recognize illegal and unsafe content in Chinese contexts. In this work, we present a Chinese safety benchmark (ChineseSafe) to facilitate research on the content safety of large language models. To align with the regulations for Chinese Internet content moderation, our ChineseSafe contains 205,034 examples across 4 classes and 10 sub-classes of safety issues. For Chinese contexts, we add several special types of illegal content: political sensitivity, pornography, and variant/homophonic words. Moreover, we employ two methods to evaluate the legal risks of popular LLMs, including open-sourced models and APIs. The results reveal that many LLMs exhibit vulnerability to certain types of safety issues, leading to legal risks in China. Our work provides a guideline for developers and researchers to facilitate the safety of LLMs. Our results are also available at https://huggingface.co/spaces/SUSTech/ChineseSafe-Benchmark.

Fine-tuning can Help Detect Pretraining Data from Large Language Models

In the era of large language models (LLMs), detecting pretraining data has been increasingly important due to concerns about fair evaluation and ethical risks. Current methods differentiate members and non-members by designing scoring functions, like Perplexity and Min-k%. However, the diversity and complexity of training data magnifies the difficulty of distinguishing, leading to suboptimal performance in detecting pretraining data. In this paper, we first explore the benefits of unseen data, which can be easily collected after the release of the LLM. We find that the perplexities of LLMs perform differently for members and non-members, after fine-tuning with a small amount of previously unseen data. In light of this, we introduce a novel and effective method termed Fine-tuned Score Deviation (FSD), which improves the performance of current scoring functions for pretraining data detection. In particular, we propose to measure the deviation distance of current scores after fine-tuning on a small amount of unseen data within the same domain. In effect, using a few unseen data can largely decrease the scores of all non-members, leading to a larger deviation distance than members. Extensive experiments demonstrate the effectiveness of our method, significantly improving the AUC score on common benchmark datasets across various models.

Knowledge Distillation with Multi-granularity Mixture of Priors for Image Super-Resolution

Knowledge distillation (KD) is a promising yet challenging model compression technique that transfers rich learning representations from a well-performing but cumbersome teacher model to a compact student model. Previous methods for image super-resolution (SR) mostly compare the feature maps directly or after standardizing the dimensions with basic algebraic operations (e.g. average, dot-product). However, the intrinsic semantic differences among feature maps are overlooked, which are caused by the disparate expressive capacity between the networks. This work presents MiPKD, a multi-granularity mixture of prior KD framework, to facilitate efficient SR model through the feature mixture in a unified latent space and stochastic network block mixture. Extensive experiments demonstrate the effectiveness of the proposed MiPKD method.