Juncheng Liu

Lichen Pan, Juncheng Liu, Yongquan Fu et al.

Distributed deep neural network training necessitates efficient GPU collective communications, which are inherently susceptible to deadlocks. GPU collective deadlocks arise easily in distributed deep learning applications when multiple collectives circularly wait for each other. GPU collective deadlocks pose a significant challenge to the correct functioning and efficiency of distributed deep learning, and no general effective solutions are currently available. Only in specific scenarios, ad-hoc methods, making an application invoke collectives in a consistent order across GPUs, can be used to prevent circular collective dependency and deadlocks. This paper presents DFCCL, a novel GPU collective communication library that provides a comprehensive approach for GPU collective deadlock prevention while maintaining high performance. DFCCL achieves preemption for GPU collectives at the bottom library level, effectively preventing deadlocks even if applications cause circular collective dependency. DFCCL ensures high performance with its execution and scheduling methods for collectives. Experiments show that DFCCL effectively prevents GPU collective deadlocks in various situations. Moreover, extensive evaluations demonstrate that DFCCL delivers performance comparable to or superior to NCCL, the state-of-the-art collective communication library highly optimized for NVIDIA GPUs.

Qian Li, Yuxiao Hu, Juncheng Liu et al.

Symbolic Regression (SR) tries to reveal the hidden equations behind observed data. However, most methods search within a discrete equation space, where the structural modifications of equations rarely align with their numerical behavior, leaving fitting error feedback too noisy to guide exploration. To address this challenge, we propose GenSR, a generative latent space-based SR framework following the `map construction -> coarse localization -> fine search'' paradigm. Specifically, GenSR first pretrains a dual-branch Conditional Variational Autoencoder (CVAE) to reparameterize symbolic equations into a generative latent space with symbolic continuity and local numerical smoothness. This space can be regarded as a well-structured `map'' of the equation space, providing directional signals for search. At inference, the CVAE coarsely localizes the input data to promising regions in the latent space. Then, a modified CMA-ES refines the candidate region, leveraging smooth latent gradients. From a Bayesian perspective, GenSR reframes the SR task as maximizing the conditional distribution $p(\mathrm{Equ.} \mid \mathrm{Num.})$, with CVAE training achieving this objective through the Evidence Lower Bound (ELBO). This new perspective provides a theoretical guarantee for the effectiveness of GenSR. Extensive experiments show that GenSR jointly optimizes predictive accuracy, expression simplicity, and computational efficiency, while remaining robust under noise.

Yufei He, Juncheng Liu, Yue Liu et al.

A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time, often behaving like "clever but clueless interns" in novel environments. This severely limits their practical utility. To systematically measure and drive progress on this challenge, we first introduce the Jericho Test-Time Learning (J-TTL) benchmark. J-TTL is a new evaluation setup where an agent must play the same game for several consecutive episodes, attempting to improve its performance from one episode to the next. On J-TTL, we find that existing adaptation methods like reflection, memory, or reinforcement learning struggle. To address the challenges posed by our benchmark, we present EvoTest, an evolutionary test-time learning framework that improves an agent without any fine-tuning or gradients-by evolving the entire agentic system after every episode. EvoTest has two roles: the Actor Agent, which plays the game, and the Evolver Agent, which analyzes the episode transcript to propose a revised configuration for the next run. This configuration rewrites the prompt, updates memory by logging effective state-action choices, tunes hyperparameters, and learns the tool-use routines. On our J-TTL benchmark, EvoTest consistently increases performance, outperforming not only reflection and memory-only baselines but also more complex online fine-tuning methods. Notably, our method is the only one capable of winning two games (Detective and Library), while all baselines fail to win any.

Kuntai Cai, Juncheng Liu, Xianglin Yang et al.

Large language model (LLM) agents typically receive two kinds of context: (i) environment-level manuals that define interaction interfaces and global rules, and (ii) task-level guidance or demonstrations tied to specific goals. In this work, we identify a crucial but overlooked third type of context, instance-level context, which consists of verifiable and reusable facts tied to a specific environment instance, such as object locations, crafting recipes, and local rules. We argue that the absence of instance-level context is a common source of failure for LLM agents in complex tasks, as success often depends not only on reasoning over global rules or task prompts but also on making decisions based on precise and persistent facts. Acquiring such context requires more than memorization: the challenge lies in efficiently exploring, validating, and formatting these facts under tight interaction budgets. We formalize this problem as Instance-Level Context Learning (ILCL) and introduce our task-agnostic method to solve it. Our method performs a guided exploration, using a compact TODO forest to intelligently prioritize its next actions and a lightweight plan-act-extract loop to execute them. This process automatically produces a high-precision context document that is reusable across many downstream tasks and agents, thereby amortizing the initial exploration cost. Experiments across TextWorld, ALFWorld, and Crafter demonstrate consistent gains in both success and efficiency: for instance, ReAct's mean success rate in TextWorld rises from 37% to 95%, while IGE improves from 81% to 95%. By transforming one-off exploration into persistent, reusable knowledge, our method complements existing contexts to enable more reliable and efficient LLM agents.