Placeholder Date
Placeholder Article Title
Placeholder excerpt text.
Our Focus
Core Research Directions
Featured Work
Research Output
Papers
Select Less, Reason More: Prioritizing Evidence Purity for Video Reasoning
Uniform frame sampling dilutes evidence in long videos, and existing pixel-space video agents lack reward mechanisms that enforce evidence purity or let them look beyond pre-sampled frames. This work introduces EARL, an evidence-aware reinforcement learning framework that turns the model into an active interrogator of evidence: it selects the most relevant frames and re-samples locally around them to recover fine-grained temporal detail. Across five video reasoning benchmarks the EARL-trained 7B model sets a new state of the art among open-source Video LLMs, reaching 59.8% on LongVideoBench, 69.0% on MVBench and 64.9% on VideoMME.
Look Less, Reason More: Rollout-Guided Adaptive Pixel-Space Reasoning
Pixel-level visual operations help Vision-Language Models handle fine-grained detail, but they are routinely overused — costing efficiency and pulling attention toward irrelevant regions. This paper proposes the first framework for adaptive pixel reasoning, which decides per query whether a pixel operation is warranted: operation-aware supervised fine-tuning establishes baseline competence, then a rollout-guided reinforcement learning stage uses feedback from the model's own responses to learn when to invoke tools. The result reaches 73.4% accuracy on HR-Bench 4K at a tool usage ratio of only 20.1% — 66.5% less tool use than prior methods, with higher accuracy.
SimRPD: Optimizing Recruitment Proactive Dialogue Agents through Simulator-Based Data Evaluation and Selection
Proactive dialogue agents steer recruitment conversations toward concrete business outcomes, but training them is bottlenecked by the scarcity of high-quality, goal-oriented domain data. SimRPD is a three-stage framework: a high-fidelity user simulator synthesizes large-scale multi-turn conversations, a Chain-of-Intention evaluation framework assesses the simulator and selects high-quality data using both global-level and instance-level metrics, and the agent is then trained on that selection. In a real-world recruitment scenario SimRPD outperforms existing simulator-based data selection strategies.
STAFDD: A Spatio-Temporal Automatic Fish Disease Detection Method
Fish disease spreads rapidly in high-density aquaculture, where manual inspection is too slow to catch it early. STAFDD combines what a fish looks like with how it moves: SSCA-YOLO, an improved YOLOv8 with an SPD-CBS downsampling module, a small-object head and coordinate attention, detects body-surface abnormalities; ByteSort with ReID features tracks individuals into stable trajectories; an LSTM reads abnormal behaviour from those trajectories; and a learnable MLP gate fuses the two branches by confidence. On a dataset built from Aeromonas hydrophila-challenged Prussian carp, tracking reaches 89.74% HOTA and 94.05% MOTA, and overall classification accuracy reaches 96.86%.
Products
Placeholder Product Name
Placeholder description text.
Placeholder Product Name
Placeholder description text.
Insights
Latest Insights
Placeholder Date
Placeholder Article Title
Placeholder excerpt text.
Placeholder Date
Placeholder Article Title
Placeholder excerpt text.