arXiv subjects

Yihong Zhang

Publications and source records attributed to Yihong Zhang.

18 recordsLinked to original sources

arXiv2026

AI Alignment through a Game-theoretic Lens: A Survey

As large language models and increasingly capable AI agents are deployed in high-risk settings, aligning them with complex human values has become a central challenge. Existing alignment methods, while effective in improving helpfulness, harmlessness, and controllability, often struggle to capture real-world preferences that are context-dependent, non-transitive, and shaped by dynamic multi-party interactions. This survey reviews AI alignment through a game-theoretic lens. Specifically, it organizes recent progress around key game-theoretic elements and synthesizes the literature along three challenges: preference diversity, alignment priority, and temporal dynamics. This perspective clarifies where current alignment methods genuinely benefit from game-theoretic analysis, where the framework is looser, and what challenges remain in building robust, adaptive, and verifiable AI systems.

cs.AI↗

arXiv2026

Rewrite System Showdown: Stochastic Search vs. EqSat

Equality saturation has become a dominant paradigm for equational program optimization. However, it has never been rigorously compared to another approach to the same problem, even though several exist, the most notable being stochastic search. In this paper, we compare equality saturation to stochastic search over five benchmarks to answer the question: are e-graphs actually good?

cs.PL↗

arXiv2026

FusionRCG: Orchestrating Recursive Computation Graphs across GPU Memory Hierarchies

Evaluating high-dimensional integrals via deep hierarchical recurrences is a dominant cost in quantum chemistry. While CPUs manage these efficiently, GPUs suffer a critical mismatch: limited per-thread memory is quickly overwhelmed by an explosion of simultaneously live intermediate variables. As recurrence scales, this forces massive data spilling to global memory, collapsing performance into a severe memory-bound regime. We present FusionRCG, a framework that jointly optimizes computation graph structure and GPU memory mapping. Exploiting the inherent topological flexibility of recurrence graphs, using electron repulsion integrals as an example, we contribute: (1) liveness-aware graph orchestration to minimize peak live intermediates; (2) algebraic dimensionality reduction via stepwise Cartesian-to-spherical fusion, shrinking intermediate footprints by up to $7.7\times$; and (3) an adaptive multi-tier kernel architecture routing graphs across the memory hierarchy. Evaluated on NVIDIA A100 GPUs, FusionRCG achieves up to $3.09\times$ end-to-end SCF speedup over GPU4PySCF and maintains $75\%$ parallel efficiency at 64~GPUs, successfully rescuing these workloads from memory-bound limits.

physics.comp-ph↗

arXiv2026

Accelerating Locality-Driven Integration in Quantum Chemistry with Block-Structured Matrix Multiplication

Locality-driven integration is a pervasive computational pattern in quantum chemistry, arising whenever spatially localized basis functions interact through numerical quadrature or integral screening. The dominant matrix multiplications in these tasks exhibit dynamic, structured sparsity driven by spatial locality, posing significant challenges for both dense batched kernels and generic sparse formats on GPUs. We present KerneLDI, a GPU-oriented framework that addresses this regime by co-designing data layout, screening logic, and matrix-computation operators to realize block-structured matrix multiplication for locality-driven integration. KerneLDI reorganizes operand matrices into a unified block-filtered representation that retains only spatially relevant blocks, and executes the resulting contractions with customized dense block multipliers that adapt proven dense-matmul optimizations to retained block pairs. We develop and evaluate KerneLDI on exchange--correlation (EXC) integration in Kohn--Sham density functional theory, a representative and computationally critical instance of this pattern. Across diverse molecular systems, KerneLDI preserves numerical accuracy while delivering up to 10$\times$ speedup for EXC evaluation over a dense GPU baseline, scales favorably with increasing system size and multi-GPU parallelism, accelerates end-to-end self-consistent field calculations, and yields nearly 6$\times$ throughput improvement for ab initio molecular dynamics.

physics.comp-ph↗

arXiv2026

Pushing Tensor Accelerators Beyond MatMul in a User-Schedulable Language

Tensor accelerators now represent a growing share of compute resources in modern CPUs and GPUs. However, they are hard to program, leading developers to use vendor-provided kernel libraries that support tensor accelerators. As a result, the usage of tensor accelerators is limited to the provided interface, mainly designed for traditional ML and scientific computing workloads. In this paper, we show that tensor accelerators can improve the performance of applications beyond simple variants of MatMul. For example, many image processing pipelines are linear transformations over matrices in disguise and can therefore utilize such specialized hardware. This is nonetheless hindered by the difficulties in programming tensor accelerators. We tackle this problem with compiler-based techniques. We use the Halide user-schedulable language and express operations as Halide algorithms succinctly. To this end, we implement a flexible tensor instruction selector based on equality saturation. The tensor instruction selector supports both CPU- and GPU-attached tensor accelerators and works with existing scheduling operations (e.g., producer-consumer fusion). Together, this enables developers to write diverse accelerator-leveraging applications in a few dozen lines. Using our system, we demonstrate the potential of tensor accelerators beyond their traditional domains. We implement several image processing pipelines (e.g., filtering, resampling, and denoising) in our system and evaluate them against non-accelerator-leveraging baselines. We show that these pipelines can achieve significant speedups. For example, a downsampling routine is sped up by $6.1\times$ by utilizing Tensor Cores on an Nvidia RTX 4070 GPU.

cs.PL↗

arXiv2025

Semantic foundations of equality saturation

Equality saturation is an emerging technique for program and query optimization developed in the programming language community. It performs term rewriting over an E-graph, a data structure that compactly represents a program space. Despite its popularity, the theory of equality saturation lags behind the practice. In this paper, we define a fixpoint semantics of equality saturation based on tree automata and uncover deep connections between equality saturation and the chase. We characterize the class of chase sequences that correspond to equality saturation. We study the complexities of terminations of equality saturation in three cases: single-instance, all-term-instance, and all-E-graph-instance. Finally, we define a syntactic criterion based on acyclicity that implies equality saturation termination.

cs.PL↗

arXiv2024

E-Graphs as Circuits, and Optimal Extraction via Treewidth

We demonstrate a new connection between e-graphs and Boolean circuits. This allows us to adapt existing literature on circuits to easily arrive at an algorithm for optimal e-graph extraction, parameterized by treewidth, which runs in $2^{O(w^2)}\text{poly}(w, n)$ time, where $w$ is the treewidth of the e-graph. Additionally, we show how the circuit view of e-graphs allows us to apply powerful simplification techniques, and we analyze a dataset of e-graphs to show that these techniques can reduce e-graph size and treewidth by 40-80% in many cases. While the core parameterized algorithm may be adapted to work directly on e-graphs, the primary value of the circuit view is in allowing the transfer of ideas from the well-established field of circuits to e-graphs.

cs.DS↗

arXiv2023

Efficient Bottom-Up Synthesis for Programs with Local Variables

We propose a new synthesis algorithm that can efficiently search programs with local variables (e.g., those introduced by lambdas). Prior bottom-up synthesis algorithms are not able to evaluate programs with free local variables, and therefore cannot effectively reduce the search space of such programs (e.g., using standard observational equivalence reduction techniques), making synthesis slow. Our algorithm can reduce the space of programs with local variables. The key idea, dubbed lifted interpretation, is to lift up the program interpretation process, from evaluating one program at a time to simultaneously evaluating all programs from a grammar. Lifted interpretation provides a mechanism to systematically enumerate all binding contexts for local variables, thereby enabling us to evaluate and reduce the space of programs with local variables. Our ideas are instantiated in the domain of web automation. The resulting tool, Arborist, can automate a significantly broader range of challenging tasks more efficiently than state-of-the-art techniques including WebRobot and Helena.

cs.PL↗

arXiv2023

Organized Event Participant Prediction Enhanced by Social Media Retweeting Data

Nowadays, many platforms on the Web offer organized events, allowing users to be organizers or participants. For such platforms, it is beneficial to predict potential event participants. Existing work on this problem tends to borrow recommendation techniques. However, compared to e-commerce items and purchases, events and participation are usually of a much smaller frequency, and the data may be insufficient to learn an accurate model. In this paper, we propose to utilize social media retweeting activity data to enhance the learning of event participant prediction models. We create a joint knowledge graph to bridge the social media and the target domain, assuming that event descriptions and tweets are written in the same language. Furthermore, we propose a learning model that utilizes retweeting information for the target domain prediction more effectively. We conduct comprehensive experiments in two scenarios with real-world data. In each scenario, we set up training data of different sizes, as well as warm and cold test cases. The evaluation results show that our approach consistently outperforms several baseline models, especially with the warm test cases, and when target domain data is limited.

cs.LG↗

arXiv2023

Better Together: Unifying Datalog and Equality Saturation

We present egglog, a fixpoint reasoning system that unifies Datalog and equality saturation (EqSat). Like Datalog, it supports efficient incremental execution, cooperating analyses, and lattice-based reasoning. Like EqSat, it supports term rewriting, efficient congruence closure, and extraction of optimized terms. We identify two recent applications--a unification-based pointer analysis in Datalog and an EqSat-based floating-point term rewriter--that have been hampered by features missing from Datalog but found in EqSat or vice-versa. We evaluate egglog by reimplementing those projects in egglog. The resulting systems in egglog are faster, simpler, and fix bugs found in the original systems.

cs.PL↗

arXiv2023

Causal Disentangled Variational Auto-Encoder for Preference Understanding in Recommendation

Recommendation models are typically trained on observational user interaction data, but the interactions between latent factors in users' decision-making processes lead to complex and entangled data. Disentangling these latent factors to uncover their underlying representation can improve the robustness, interpretability, and controllability of recommendation models. This paper introduces the Causal Disentangled Variational Auto-Encoder (CaD-VAE), a novel approach for learning causal disentangled representations from interaction data in recommender systems. The CaD-VAE method considers the causal relationships between semantically related factors in real-world recommendation scenarios, rather than enforcing independence as in existing disentanglement methods. The approach utilizes structural causal models to generate causal representations that describe the causal relationship between latent factors. The results demonstrate that CaD-VAE outperforms existing methods, offering a promising solution for disentangling complex user behavior data in recommendation systems.

cs.IR↗

arXiv2023

Scalable fast benchmarking for individual quantum gates with local twirling

With the development of controllable quantum systems, fast and practical characterization for multi-qubit gates is essential for building high-fidelity quantum computing devices. The usual way to fulfill this requirement via randomized benchmarking asks for the complicated implementation of numerous multi-qubit twirling gates. How to efficiently and reliably estimate the fidelity of a quantum process remains an open problem. In this work, we propose a character-cycle benchmarking protocol and a character-average benchmarking protocol only using local twirling gates to estimate the process fidelity of an individual multi-qubit operation. Our protocols can characterize a large class of quantum gates including and beyond the Clifford group via the local gauge transformation, which forms a universal gate set for quantum computing. We numerically demonstrate our protocols for a non-Clifford gate -- controlled-$(TX)$ and a Clifford gate -- five-qubit quantum error-correcting encoding circuit. The numerical results show that our protocols can efficiently and reliably characterize the gate process fidelities. Compared with the cross-entropy benchmarking, the simulation results show that the character-average benchmarking achieves three orders of magnitude improvements in terms of sampling complexity.

quant-ph↗

arXiv2022

Debiasing Graph Transfer Learning via Item Semantic Clustering for Cross-Domain Recommendations

Deep learning-based recommender systems may lead to over-fitting when lacking training interaction data. This over-fitting significantly degrades recommendation performances. To address this data sparsity problem, cross-domain recommender systems (CDRSs) exploit the data from an auxiliary source domain to facilitate the recommendation on the sparse target domain. Most existing CDRSs rely on overlapping users or items to connect domains and transfer knowledge. However, matching users is an arduous task and may involve privacy issues when data comes from different companies, resulting in a limited application for the above CDRSs. Some studies develop CDRSs that require no overlapping users and items by transferring learned user interaction patterns. However, they ignore the bias in user interaction patterns between domains and hence suffer from an inferior performance compared with single-domain recommender systems. In this paper, based on the above findings, we propose a novel CDRS, namely semantic clustering enhanced debiasing graph neural recommender system (SCDGN), that requires no overlapping users and items and can handle the domain bias. More precisely, SCDGN semantically clusters items from both domains and constructs a cross-domain bipartite graph generated from item clusters and users. Then, the knowledge is transferred via this cross-domain user-cluster graph from the source to the target. Furthermore, we design a debiasing graph convolutional layer for SCDGN to extract unbiased structural knowledge from the cross-domain user-cluster graph. Our Experimental results on three public datasets and a pair of proprietary datasets verify the effectiveness of SCDGN over state-of-the-art models in terms of cross-domain recommendations.

cs.IR↗

arXiv2022

Relational E-Matching

We present a new approach to e-matching based on relational join; in particular, we apply recent database query execution techniques to guarantee worst-case optimal run time. Compared to the conventional backtracking approach that always searches the e-graph "top down", our new relational e-matching approach can better exploit pattern structure by searching the e-graph according to an optimized query plan. We also establish the first data complexity result for e-matching, bounding run time as a function of the e-graph size and output size. We prototyped and evaluated our technique in the state-of-the-art egg e-graph framework. Compared to a conventional baseline, relational e-matching is simpler to implement and orders of magnitude faster in practice.

cs.DB↗

arXiv2021

A General Method for Event Detection on Social Media

Event detection on social media has attracted a number of researches, given the recent availability of large volumes of social media discussions. Previous works on social media event detection either assume a specific type of event, or assume certain behavior of observed variables. In this paper, we propose a general method for event detection on social media that makes few assumptions. The main assumption we make is that when an event occurs, affected semantic aspects will behave differently from its usual behavior. We generalize the representation of time units based on word embeddings of social media text, and propose an algorithm to detect events in time series in a general sense. In the experimental evaluation, we use a novel setting to test if our method and baseline methods can exhaustively catch all real-world news in the test period. The evaluation results show that when the event is quite unusual with regard to the base social media discussion, it can be captured more effectively with our method. Our method can be easily implemented and can be treated as a starting point for more specific applications.

cs.IR↗

arXiv2021

Using Social Media Background to Improve Cold-start Recommendation Deep Models

In recommender systems, a cold-start problem occurs when there is no past interaction record associated with the user or item. Typical solutions to the cold-start problem make use of contextual information, such as user demographic attributes or product descriptions. A group of works have shown that social media background can help predicting temporal phenomenons such as product sales and stock price movements. In this work, our goal is to investigate whether social media background can be used as extra contextual information to improve recommendation models. Based on an existing deep neural network model, we proposed a method to represent temporal social media background as embeddings and fuse them as an extra component in the model. We conduct experimental evaluations on a real-world e-commerce dataset and a Twitter dataset. The results show that our method of fusing social media background with the existing model does generally improve recommendation performance. In some cases the recommendation accuracy measured by hit-rate@K doubles after fusing with social media background. Our findings can be beneficial for future recommender system designs that consider complex temporal information representing social interests.

cs.IR↗

arXiv2021

Efficient entanglement generation and detection of generalized stabilizer states

The generation and verification of large-scale entanglement are essential to the development of quantum technologies. In this paper, we present an efficient scheme to generate genuine multipartite entanglement of a large number of qubits by using the Heisenberg interaction. This method can be conveniently implemented in various physical platforms, including superconducting, trapped-ion, and cold-atom systems. In order to characterize the entanglement of the output quantum state, we generalize the stabilizer formalism and develop an entanglement witness method. In particular, we design a generic searching algorithm to optimize entanglement witness with a minimal number of measurement settings under a given noise level. From the perspective of practical applications, we numerically study the trade-off between the experiment efficiency and the detection robustness.

quant-ph↗

arXiv2021

GeCo: Quality Counterfactual Explanations in Real Time

Machine learning is increasingly applied in high-stakes decision making that directly affect people's lives, and this leads to an increased demand for systems to explain their decisions. Explanations often take the form of counterfactuals, which consists of conveying to the end user what she/he needs to change in order to improve the outcome. Computing counterfactual explanations is challenging, because of the inherent tension between a rich semantics of the domain, and the need for real time response. In this paper we present GeCo, the first system that can compute plausible and feasible counterfactual explanations in real time. At its core, GeCo relies on a genetic algorithm, which is customized to favor searching counterfactual explanations with the smallest number of changes. To achieve real-time performance, we introduce two novel optimizations: $Δ$-representation of candidate counterfactuals, and partial evaluation of the classifier. We compare empirically GeCo against five other systems described in the literature, and show that it is the only system that can achieve both high quality explanations and real time answers.

cs.LG↗