Pith. sign in

REVIEW 6 cited by

HLM-Cite: Hybrid Language Model Workflow for Text-based Scientific Citation Prediction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2410.09112 v1 pith:V46WF5RA submitted 2024-10-10 cs.DL cs.AIcs.CL

classification cs.DLcs.AIcs.CL
keywords citationcitationsrelationshipstextbfcorecriticalhlm-citelogical
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Citation networks are critical in modern science, and predicting which previous papers (candidates) will a new paper (query) cite is a critical problem. However, the roles of a paper's citations vary significantly, ranging from foundational knowledge basis to superficial contexts. Distinguishing these roles requires a deeper understanding of the logical relationships among papers, beyond simple edges in citation networks. The emergence of LLMs with textual reasoning capabilities offers new possibilities for discerning these relationships, but there are two major challenges. First, in practice, a new paper may select its citations from gigantic existing papers, where the texts exceed the context length of LLMs. Second, logical relationships between papers are implicit, and directly prompting an LLM to predict citations may result in surface-level textual similarities rather than the deeper logical reasoning. In this paper, we introduce the novel concept of core citation, which identifies the critical references that go beyond superficial mentions. Thereby, we elevate the citation prediction task from a simple binary classification to distinguishing core citations from both superficial citations and non-citations. To address this, we propose $\textbf{HLM-Cite}$, a $\textbf{H}$ybrid $\textbf{L}$anguage $\textbf{M}$odel workflow for citation prediction, which combines embedding and generative LMs. We design a curriculum finetune procedure to adapt a pretrained text embedding model to coarsely retrieve high-likelihood core citations from vast candidates and then design an LLM agentic workflow to rank the retrieved papers through one-shot reasoning, revealing the implicit relationships among papers. With the pipeline, we can scale the candidate sets to 100K papers. We evaluate HLM-Cite across 19 scientific fields, demonstrating a 17.6% performance improvement comparing SOTA methods.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. THE-Tree: Can Tracing Historical Evolution Enhance Scientific Verification and Reasoning?

    cs.AI 2025-06 reject novelty 6.0 of 10

    THE-Tree constructs causally-linked semantic evolution trees from surveys and literature, and the authors report improved graph completion, future prediction, and LLM-based paper evaluation.

  2. Reason-to-Recommend: Using Interaction-of-Thought Reasoning to Enhance LLM Recommendation

    cs.IR 2025-06 conditional novelty 6.0 of 10

    R2Rec combines graph-sampled interaction chains, masked stepwise reasoning prompts, and SFT plus GRPO training to improve LLM recommendation hit ratios.

  3. Open-Set Living Need Prediction with Large Language Models

    cs.AI 2025-06 conditional novelty 6.0 of 10

    PIGEON uses LLMs with retrieved user history and Maslow's hierarchy to predict open-set living needs in free text, improving life service recall over closed-set baselines.

  4. Multiple Weaks Win Single Strong: Large Language Models Ensemble Weak Reinforcement Learning Agents into a Supreme One

    cs.LG 2025-05 reject novelty 6.0 of 10

    An LLM classifies game states into situations and selects the RL agent with the best historical average reward for each situation, outperforming static ensemble baselines on Atari.

  5. AI4Research: A Survey of Artificial Intelligence for Scientific Research

    cs.CL 2025-07 conditional novelty 4.0 of 10

    A survey that organizes AI-for-research work into five tasks, comprehension, survey, discovery, writing, and peer review, and compiles associated tools and benchmarks.

  6. Federated Learning-Enabled Hybrid Language Models for Communication-Efficient Token Transmission

    cs.LG 2025-06 reject novelty 4.0 of 10

    FedHLM uses federated learning to learn token-level uncertainty thresholds that decide when to offload tokens from a small edge LM to a large cloud LM, claiming a 95 percent reduction in LLM transmissions.

Pith tools