Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
125 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Large Language Models for Social Networks: Applications, Challenges, and Solutions (2401.02575v1)

Published 4 Jan 2024 in cs.SI, cs.AI, and cs.LG

Abstract: LLMs are transforming the way people generate, explore, and engage with content. We study how we can develop LLM applications for online social networks. Despite LLMs' successes in other domains, it is challenging to develop LLM-based products for social networks for numerous reasons, and it has been relatively under-reported in the research community. We categorize LLM applications for social networks into three categories. First is knowledge tasks where users want to find new knowledge and information, such as search and question-answering. Second is entertainment tasks where users want to consume interesting content, such as getting entertaining notification content. Third is foundational tasks that need to be done to moderate and operate the social networks, such as content annotation and LLM monitoring. For each task, we share the challenges we found, solutions we developed, and lessons we learned. To the best of our knowledge, this is the first comprehensive paper about developing LLM applications for social networks.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (93)
  1. Self-rag: Learning to retrieve, generate, and critique through self-reflection. arXiv preprint arXiv:2310.11511.
  2. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862.
  3. The reversal curse: Llms trained on" a is b" fail to learn" b is a". arXiv preprint arXiv:2309.12288.
  4. Safety-tuned llamas: Lessons from improving the safety of large language models that follow instructions. arXiv preprint arXiv:2309.07875.
  5. Improving language models by retrieving from trillions of tokens. In International conference on machine learning, pages 2206–2240. PMLR.
  6. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901.
  7. Training language models with language feedback. In ACL Workshop on Learning with Natural Language Supervision. 2022.
  8. Open problems and fundamental limitations of reinforcement learning from human feedback. arXiv preprint arXiv:2307.15217.
  9. Chateval: Towards better llm-based evaluators through multi-agent debate. arXiv preprint arXiv:2308.07201.
  10. Understanding retrieval augmentation for long-form question answering. arXiv preprint arXiv:2310.12150.
  11. Rich knowledge sources bring complex knowledge conflicts: Recalibrating models to reflect conflicting evidence. arXiv preprint arXiv:2210.13701.
  12. Adapting large language models via reading comprehension. arXiv preprint arXiv:2309.09530.
  13. Factool: Factuality detection in generative ai–a tool augmented framework for multi-task and multi-domain scenarios. arXiv preprint arXiv:2307.13528.
  14. A closer look into automatic evaluation using large language models. arXiv preprint arXiv:2310.05657.
  15. Dola: Decoding by contrasting layers improves factuality in large language models. arXiv preprint arXiv:2309.03883.
  16. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416.
  17. Chain-of-verification reduces hallucination in large language models. arXiv preprint arXiv:2309.11495.
  18. Gptscore: Evaluate as you desire. arXiv preprint arXiv:2302.04166.
  19. Google (2023). Bard: A conversational ai tool by google. https://bard.google.com/.
  20. Don’t stop pretraining: Adapt language models to domains and tasks. arXiv preprint arXiv:2004.10964.
  21. Retrieval augmented language model pre-training. In International conference on machine learning, pages 3929–3938. PMLR.
  22. Inspecting and editing knowledge representations in language models. arXiv preprint arXiv:2304.00740.
  23. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531.
  24. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685.
  25. Lawyer llama technical report. arXiv preprint arXiv:2305.15062.
  26. Efficient long-text understanding with short-text models. Transactions of the Association for Computational Linguistics, 11:284–299.
  27. Few-shot learning with retrieval augmented language models. arXiv preprint arXiv:2208.03299.
  28. Retrieval as attention: End-to-end learning of retrieval and reading within a single transformer. arXiv preprint arXiv:2212.02027.
  29. Active retrieval augmented generation. arXiv preprint arXiv:2305.06983.
  30. Learning, S.-S. (2006). Semi-supervised learning. CSZ2006. html.
  31. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems, 33:9459–9474.
  32. Prd: Peer rank and discussion improve large language model based evaluations. arXiv preprint arXiv:2307.02762.
  33. Lin, C.-Y. (2004). Rouge: A package for automatic evaluation of summaries. In Text summarization branches out, pages 74–81.
  34. Lost in the middle: How language models use long contexts. arXiv preprint arXiv:2307.03172.
  35. In-context vectors: Making in context learning more effective and controllable through latent space steering. arXiv preprint arXiv:2311.06668.
  36. G-eval: Nlg evaluation using gpt-4 with better human alignment, may 2023. arXiv preprint arXiv:2303.16634.
  37. Gpteval: Nlg evaluation using gpt-4 with better human alignment. arXiv preprint arXiv:2303.16634.
  38. Selfcheckgpt: Zero-resource black-box hallucination detection for generative large language models. arXiv preprint arXiv:2303.08896.
  39. Locating and editing factual associations in gpt. Advances in Neural Information Processing Systems, 35:17359–17372.
  40. Can a suit of armor conduct electricity? a new dataset for open book question answering. arXiv preprint arXiv:1809.02789.
  41. Factscore: Fine-grained atomic evaluation of factual precision in long form text generation. arXiv preprint arXiv:2305.14251.
  42. An emulator for fine-tuning large language models using small language models. arXiv preprint arXiv:2310.12962.
  43. Few-shot fine-tuning vs. in-context learning: A fair comparison and evaluation. arXiv preprint arXiv:2305.16938.
  44. OpenAI (2023a). Gpt-3.5 turbo fine-tuning and api updates. https://openai.com/blog/gpt-3-5-turbo-fine-tuning-and-api-updates.
  45. OpenAI (2023b). Gpt-4 technical report.
  46. OpenAI (2023c). Moderation - openai api. https://platform.openai.com/docs/guides/moderation.
  47. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730–27744.
  48. Fine-tuning or retrieval? comparing knowledge injection in llms.
  49. Hyena hierarchy: Towards larger convolutional language models. arXiv preprint arXiv:2302.10866.
  50. Fine-tuning aligned language models compromises safety, even when users do not intend to! arXiv preprint arXiv:2310.03693.
  51. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9.
  52. Direct preference optimization: Your language model is secretly a reward model. arXiv preprint arXiv:2305.18290.
  53. Squad: 100,000+ questions for machine comprehension of text. arXiv preprint arXiv:1606.05250.
  54. In-context retrieval-augmented language models. arXiv preprint arXiv:2302.00083.
  55. Social bias frames: Reasoning about social and power implications of language. arXiv preprint arXiv:1911.03891.
  56. Training language models with language feedback at scale. arXiv preprint arXiv:2303.16755.
  57. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347.
  58. Wikichat: A few-shot llm-based chatbot grounded with wikipedia. arXiv preprint arXiv:2305.14292.
  59. Large language models can be easily distracted by irrelevant context. In International Conference on Machine Learning, pages 31210–31227. PMLR.
  60. Large language models encode clinical knowledge. arXiv preprint arXiv:2212.13138.
  61. Towards expert-level medical question answering with large language models. arXiv preprint arXiv:2305.09617.
  62. Commonsenseqa: A question answering challenge targeting commonsense knowledge. arXiv preprint arXiv:1811.00937.
  63. Alpaca: A strong, replicable instruction-following model. Stanford Center for Research on Foundation Models. https://crfm. stanford. edu/2023/03/13/alpaca. html, 3(6):7.
  64. Fine-tuning language models for factuality.
  65. Fine-tuning language models for factuality. arXiv preprint arXiv:2311.08401.
  66. Just ask for calibration: Strategies for eliciting calibrated confidence scores from language models fine-tuned with human feedback. arXiv preprint arXiv:2305.14975.
  67. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288.
  68. Interleaving retrieval with chain-of-thought reasoning for knowledge-intensive multi-step questions. arXiv preprint arXiv:2212.10509.
  69. Attention is all you need. Advances in neural information processing systems, 30.
  70. Freshllms: Refreshing large language models with search engine augmentation. arXiv preprint arXiv:2310.03214.
  71. Is chatgpt a good nlg evaluator? a preliminary study. arXiv preprint arXiv:2303.04048.
  72. Large search model: Redefining search stack in the era of llms. arXiv preprint arXiv:2310.14587.
  73. Self-instruct: Aligning language model with self generated instructions. arXiv preprint arXiv:2212.10560.
  74. Finetuned language models are zero-shot learners. arXiv preprint arXiv:2109.01652.
  75. System 2 attention (is something you might need too). arXiv preprint arXiv:2311.11829.
  76. Bloomberggpt: A large language model for finance. arXiv preprint arXiv:2303.17564.
  77. An explanation of in-context learning as implicit bayesian inference. arXiv preprint arXiv:2111.02080.
  78. Doctorglm: Fine-tuning your chinese doctor is not a herculean task. arXiv preprint arXiv:2304.01097.
  79. Pretraining data mixtures enable narrow model selection capabilities in transformer models. arXiv preprint arXiv:2311.00871.
  80. React: Synergizing reasoning and acting in language models. arXiv preprint arXiv:2210.03629.
  81. Making retrieval-augmented language models robust to irrelevant context. arXiv preprint arXiv:2310.01558.
  82. Chatdoctor: A medical chat model fine-tuned on llama model using medical domain knowledge. arXiv preprint arXiv:2303.14070.
  83. Let ai entertain you: Increasing user engagement with generative ai and rejection sampling.
  84. Evaluating large language models at evaluating instruction following. arXiv preprint arXiv:2310.07641.
  85. How language model hallucinations can snowball. arXiv preprint arXiv:2305.13534.
  86. Adaptive budget allocation for parameter-efficient fine-tuning. arXiv preprint arXiv:2303.10512.
  87. Tell your model where to attend: Post-hoc attention steering for llms. arXiv preprint arXiv:2311.02262.
  88. Llama-adapter: Efficient fine-tuning of language models with zero-init attention. arXiv preprint arXiv:2303.16199.
  89. Wider and deeper llm networks are fairer llm evaluators. arXiv preprint arXiv:2308.01862.
  90. Knowledgeable preference alignment for llms in domain-specific question answering. arXiv preprint arXiv:2311.06503.
  91. Domain specialization as the key to make large language models disruptive: A comprehensive survey. arXiv preprint arXiv:2305.18703.
  92. Slic-hf: Sequence likelihood calibration with human feedback. arXiv preprint arXiv:2305.10425.
  93. Judging llm-as-a-judge with mt-bench and chatbot arena. arXiv preprint arXiv:2306.05685.
Citations (3)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now