paper-with-me

홈 › Papers

DeepInnovator: Triggering the Innovative Capabilities of LLMs

2026-02-21 · Tianyu Fan, Fengji Zhang, Yuxiang Zheng, Bei Chen, Xinyao Niu, Chengen Huang, Junyang Lin, Chao Huang arxiv

The application of Large Language Models (LLMs) in accelerating scientific discovery has garnered increasing attention, with a key focus on constructing research agents endowed with innovative capability, i.e., the ability to autonomously generate novel and significant research ideas. Existing approaches predominantly rely on sophisticated prompt engineering and lack a systematic training paradigm. To address this, we propose DeepInnovator, a training framework designed to trigger the innovative capability of LLMs. Our approach comprises two core components. (1) `Standing on the shoulders of giants''. We construct an automated data extraction pipeline to extract and organize structured research knowledge from a vast corpus of unlabeled scientific literature. (2) Conjectures and refutations''. We introduce a `Next Idea Prediction'' training paradigm, which models the generation of research ideas as an iterative process of continuously predicting, evaluating, and refining plausible and novel next idea. Both automatic and expert evaluations demonstrate that our DeepInnovator-14B significantly outperforms untrained baselines, achieving win rates of 80.53\%-93.81\%, and attains performance comparable to that of current leading LLMs. This work provides a scalable training pathway toward building research agents with genuine, originative innovative capability, and will open-source the dataset to foster community advancement. Source code and data are available at: https://github.com/HKUDS/DeepInnovator.

📄 PDF Abstract BibTeX arXiv:2602.18920

Code (0)

등록된 구현이 없습니다.

Tasks

Prompt Engineering

Similar Papers 제목 키워드 기반

AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning

2025-05-17 · Chenwei Lou, Zewei Sun, Xinnian Liang, Meng Qu 외

Large Language Models (LLMs) have demonstrated remarkable capabilities but often face challenges with tasks requiring sophisticated reasoning. While Chain-of-Thought (CoT) prompting significantly enhances reasoning, it i…

Reinforcement Learning (RL)

High-Fidelity Event-Radiance Recovery via Transient Event Frequency

2023-01-01 · CVPR 2023 1 · Jin Han, Yuta Asano, Boxin Shi, Yinqiang Zheng 외

High-fidelity radiance recovery plays a crucial role in scene information reconstruction and understanding. Conventional cameras suffer from limited sensitivity in dynamic range, bit depth, and spectral response, etc…

Sensitivity

SAEExplainer: Interpreting SAE Features with Activation-Guided Preference Optimization

2026-06-07 · Jingyi He, Haiyan Zhao, Ruxue Shi, Yanguang Liu 외 arxiv

Although Sparse Autoencoders (SAEs) have mitigated the opacity of large language models (LLMs) by decomposing dense representations into sparse features, explaining these features still remains a central challenge. Curre…

Re-Triggering Safeguards within LLMs for Jailbreak Detection

2026-05-11 · Zheng Lin, Zhenxing Niu, Haoxuan Ji, Yuzhe Huang 외 arxiv

This paper proposes a jailbreaking prompt detection method for large language models (LLMs) to defend against jailbreak attacks. Although recent LLMs are equipped with built-in safeguards, it remains possible to craft ja…

Enhancing ID-based Recommendation with Large Language Models

2024-11-04 · Lei Chen, Chen Gao, Xiaoyi Du, Hengliang Luo 외

Large Language Models (LLMs) have recently garnered significant attention in various domains, including recommendation systems. Recent research leverages the capabilities of LLMs to improve the performance and user model…

Recommendation Systems