arXiv subjects

Haibo Shi

Publications and source records attributed to Haibo Shi.

22 records · Page 2Linked to original sources

arXiv2024

Learning to Use Tools via Cooperative and Interactive Agents

Tool learning empowers large language models (LLMs) as agents to use external tools and extend their utility. Existing methods employ one single LLM-based agent to iteratively select and execute tools, thereafter incorporating execution results into the next action prediction. Despite their progress, these methods suffer from performance degradation when addressing practical tasks due to: (1) the pre-defined pipeline with restricted flexibility to calibrate incorrect actions, and (2) the struggle to adapt a general LLM-based agent to perform a variety of specialized actions. To mitigate these problems, we propose ConAgents, a Cooperative and interactive Agents framework, which coordinates three specialized agents for tool selection, tool execution, and action calibration separately. ConAgents introduces two communication protocols to enable the flexible cooperation of agents. To effectively generalize the ConAgents into open-source models, we also propose specialized action distillation, enhancing their ability to perform specialized actions in our framework. Our extensive experiments on three datasets show that the LLMs, when equipped with the ConAgents, outperform baselines with substantial improvement (i.e., up to 14% higher success rate).

cs.CL↗

arXiv2024

The Real, the Better: Aligning Large Language Models with Online Human Behaviors

Large language model alignment is widely used and studied to avoid LLM producing unhelpful and harmful responses. However, the lengthy training process and predefined preference bias hinder adaptation to online diverse human preferences. To this end, this paper proposes an alignment framework, called Reinforcement Learning with Human Behavior (RLHB), to align LLMs by directly leveraging real online human behaviors. By taking the generative adversarial framework, the generator is trained to respond following expected human behavior; while the discriminator tries to verify whether the triplets of query, response, and human behavior come from real online environments. Behavior modeling in natural-language form and the multi-model joint training mechanism enable an active and sustainable online alignment. Experimental results confirm the effectiveness of our proposed methods by both human and automatic evaluations.

cs.CL↗

arXiv2023

LOIS: Looking Out of Instance Semantics for Visual Question Answering

Visual question answering (VQA) has been intensively studied as a multimodal task that requires effort in bridging vision and language to infer answers correctly. Recent attempts have developed various attention-based modules for solving VQA tasks. However, the performance of model inference is largely bottlenecked by visual processing for semantics understanding. Most existing detection methods rely on bounding boxes, remaining a serious challenge for VQA models to understand the causal nexus of object semantics in images and correctly infer contextual information. To this end, we propose a finer model framework without bounding boxes in this work, termed Looking Out of Instance Semantics (LOIS) to tackle this important issue. LOIS enables more fine-grained feature descriptions to produce visual facts. Furthermore, to overcome the label ambiguity caused by instance masks, two types of relation attention modules: 1) intra-modality and 2) inter-modality, are devised to infer the correct answers from the different multi-view features. Specifically, we implement a mutual relation attention module to model sophisticated and deeper visual semantic relations between instance objects and background information. In addition, our proposed attention model can further analyze salient image regions by focusing on important word-related questions. Experimental results on four benchmark VQA datasets prove that our proposed method has favorable performance in improving visual reasoning capability.

cs.CV↗

arXiv2018

Research on Limited Buffer Scheduling Problems in Flexible Flow Shops with Setup Times

In order to solve the limited buffer scheduling problems in flexible flow shops with setup times, this paper proposes an improved whale optimization algorithm (IWOA) as a global optimization algorithm. Firstly, this paper presents a mathematic programming model for limited buffer in flexible flow shops with setup times, and applies the IWOA algorithm as the global optimization algorithm. Based on the whale optimization algorithm (WOA), the improved algorithm uses Levy flight, opposition-based learning strategy and simulated annealing to expand the search range, enhance the ability for jumping out of local extremum, and improve the continuous evolution of the algorithm. To verify the improvement of the proposed algorithm on the optimization ability of the standard WOA algorithm, the IWOA algorithm is tested by verification examples of small-scale and large-scale flexible flow shop scheduling problems, and the imperialist competitive algorithm (ICA), bat algorithm (BA), and whale optimization algorithm (WOA) are used for comparision. Based on the instance data of bus manufacturer, simulation tests are made on the four algorithms under variouis of practical evalucation scenarios. The simulation results show that the IWOA algorithm can better solve this type of limited buffer scheduling problem in flexible flow shops with setup times compared with the state of the art algorithms.

cs.AI↗