30 papers
Beneath the Diff: Diagnosing and Mitigating Algorithmic Mode Collapse in Code-Level Autonomous Research Loops
Bowei He, Weixu Zhang, Yili Jin +1
Code-level autonomous research loops (ARLs) have recently emerged as a concrete object of study in automated machine learning research. In such loops, an LLM agent proposes modific…
Budget-Aware LLM Discovery via Cost-Calibrated Frontier Utility
Yansen Zhang, Yilu Liu, Tianyu Liu +7
Large language models increasingly support scientific and algorithmic discovery through inference-time search over evaluated candidates. Existing adaptive discovery controllers ass…
AdaMTP: An Adaptive Training Paradigm for Multi-Token Prediction
Ziqiang Cui, Han Shi, Bowei He +8
Multi-Token Prediction (MTP) has emerged as an effective paradigm that augments a shared Large Language Model backbone with auxiliary heads, training the model to predict several f…
Beyond Document Grounding: Span-Level Hallucination Detection over Code, Tool Output, and Documents
Ádám Kovács, Bowei He, Xue Liu +3
Hallucination detection for retrieval-augmented generation (RAG) is usually evaluated on natural-language document evidence. However, grounded generation systems increasingly rely…
Awakening Dormant Experts:Counterfactual Routing to Mitigate MoE Hallucinations
Wentao Hu, Yanbo Zhai, Xiaohui Hu +6
Sparse Mixture-of-Experts (MoE) models have achieved remarkable scalability, yet they remain vulnerable to hallucinations, particularly when processing long-tail knowledge. We iden…
The 1/W Law: An Analytical Study of Context-Length Routing Topology and GPU Generation Gains for LLM Inference Energy Efficiency
Huamin Chen, Xunzhuo Liu, Yuhan Liu +3
How many tokens can a GPU inference cluster deliver per watt? Across deployments of identical hardware, the answer varies by 40x -- not because of software inefficiency, but becaus…
Seedance 2.0: Advancing Video Generation for World Complexity
Team Seedance, De Chen, Liyang Chen +168
Seedance 2.0 is a new native multi-modal audio-video generation model, officially released in China in early February 2026. Compared with its predecessors, Seedance 1.0 and 1.5 Pro…
GAM: Hierarchical Graph-based Agentic Memory for LLM Agents
Zhaofen Wu, Hanrong Zhang, Fulin Lin +9
To sustain coherent long-term interactions, Large Language Model (LLM) agents must navigate the tension between acquiring new information and retaining prior knowledge. Current uni…
Token-Budget-Aware Pool Routing for Cost-Efficient LLM Inference
Huamin Chen, Xunzhuo Liu, Junchen Jiang +2
Production vLLM fleets provision every instance for worst-case context length, wasting 4-8x concurrency on the 80-95% of requests that are short and simultaneously triggering KV-ca…
Dual-Pool Token-Budget Routing for Cost-Efficient and Reliable LLM Serving
Xunzhuo Liu, Bowei He, Xue Liu +3
Production vLLM fleets typically provision each instance for the worst-case context length, leading to substantial KV-cache over-allocation and under-utilized concurrency. In pract…
From Inference Routing to Agent Orchestration: Declarative Policy Compilation with Cross-Layer Verification
Huamin Chen, Xunzhuo Liu, Bowei He +1
The Semantic Router DSL is a non-Turing-complete policy language deployed in production for per-request LLM inference routing: content signals (embedding similarity, PII detection,…
Knowledge Access Beats Model Size: Memory Augmented Routing for Persistent AI Agents
Xunzhuo Liu, Bowei He, Xue Liu +3
Production AI agents frequently receive user-specific queries that are highly repetitive, with up to 47\% being semantically similar to prior interactions, yet each query is typica…
Conflict-Free Policy Languages for Probabilistic ML Predicates: A Framework and Case Study with the Semantic Router DSL
Xunzhuo Liu, Hao Wu, Huamin Chen +2
Conflict detection in policy languages is a solved problem -- as long as every rule condition is a crisp Boolean predicate. BDDs, SMT solvers, and NetKAT all exploit that assumptio…
FleetOpt: Analytical Fleet Provisioning for LLM Inference with Compress-and-Route as Implementation Mechanism
Huamin Chen, Xunzhuo Liu, Yuhan Liu +3
Modern LLM GPU fleets are provisioned for worst-case context lengths that the vast majority of requests never approach, wasting GPU capacity on idle KV-cache slots. We present Flee…
inference-fleet-sim: A Queueing-Theory-Grounded Fleet Capacity Planner for LLM Inference
Huamin Chen, Xunzhuo Liu, Yuhan Liu +3
Sizing a GPU fleet for LLM inference is harder than it looks. The obvious questions -- how many GPUs, which type, where to split a two-pool fleet -- have no closed-form answers. Th…
Visual Confused Deputy: Exploiting and Defending Perception Failures in Computer-Using Agents
Xunzhuo Liu, Bowei He, Xue Liu +3
Computer-using agents (CUAs) act directly on graphical user interfaces, yet their perception of the screen is often unreliable. Existing work largely treats these failures as perfo…
Outcome-Aware Tool Selection for Semantic Routers: Latency-Constrained Learning Without LLM Inference
Huamin Chen, Xunzhuo Liu, Junchen Jiang +2
Semantic routers in LLM inference gateways select tools in the critical request path, where every millisecond of added latency compounds across millions of requests. We propose Out…
Adaptive Vision-Language Model Routing for Computer Use Agents
Xunzhuo Liu, Bowei He, Xue Liu +3
Computer Use Agents (CUAs) translate natural-language instructions into Graphical User Interface (GUI) actions such as clicks, keystrokes, and scrolls by relying on a Vision-Langua…
98 Faster LLM Routing Without a Dedicated GPU: Flash Attention, Prompt Compression, and Near-Streaming for the vLLM Semantic Router
Xunzhuo Liu, Bowei He, Xue Liu +3
System-level routers that intercept LLM requests for safety classification, domain routing, and PII detection must be both fast and operationally lightweight: they should add minim…
Fast and Faithful: Real-Time Verification for Long-Document Retrieval-Augmented Generation Systems
Xunzhuo Liu, Bowei He, Xue Liu +2
Retrieval-augmented generation (RAG) is increasingly deployed in enterprise search and document-centric assistants, where responses must be grounded in long and complex source mate…
RubricBench: Aligning Model-Generated Rubrics with Human Standards
Qiyuan Zhang, Junyi Zhou, Yufei Wang +8
As Large Language Model (LLM) alignment evolves from simple completions to complex, highly sophisticated generation, Reward Models are increasingly shifting toward rubric-guided ev…
Beyond Length Scaling: Synergizing Breadth and Depth for Generative Reward Models
Qiyuan Zhang, Yufei Wang, Tianhe Wu +5
Recent advancements in Generative Reward Models (GRMs) have demonstrated that scaling the length of Chain-of-Thought (CoT) reasoning considerably enhances the reliability of evalua…
Can LLM Agents Be CFOs? Benchmarking Long-Horizon Resource Allocation in an Uncertain Enterprise Environment
Yi Han, Yan Wang, Lingfei Qian +12
Large language model (LLM) agents are increasingly tested on complex tasks, but their ability to allocate scarce resources over long horizons remains unclear. Unlike reactive tasks…
Exploring the AI Obedience: Why is Generating a Pure Color Image Harder than CyberPunk?
Hongyu Li, Kuan Liu, Yuan Chen +6
Recent advances in generative AI have shown human-level performance in complex content creation. However, we identify a "Paradox of Simplicity": models that can render complex scen…