Ir al contenido

Documat

Revisiting Challenges and Hazards in Large Language Model Evaluation

  • Autores: Íñigo López Gazpio
  • Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 72, 2024, págs. 15-30
  • Idioma: inglés
  • Títulos paralelos:
    • Análisis de los Desafíos y Riesgos en la Evaluación de Grandes Modelos del Lenguaje
  • Enlaces
  • Resumen
    • español

      En la era de los modelos de lenguaje de gran escala, el objetivo de la inteligencia artificial ha evolucionado para asistir a personas de maneras sin precedentes conocidos. A medida que los modelos se integran en la sociedad, aumenta la necesidad de evaluaciones exhaustivas. La aceptación de estos sistemas en el mundo real depende de sus habilidades de conocimiento, razonamiento y argumentación. Sin embargo, estándares inconsistentes entre dominios complican la evaluación, dificultando la comparación de modelos y la comprensión de su funcionamiento. Nuestro estudio se enfoca en organizar y aclarar los procesos de evaluación de estos modelos. Examinamos investigaciones recientes para analizar las tendencias actuales e investigar si los métodos de evaluación se ajustan a los requisitos del progreso. Finalmente, identificamos y detallamos los principales desafíos y riesgos que afectan la evaluación, un área que aún no ha sido explorada extensamente. Este enfoque es necesario para reconocer las limitaciones actuales, el potencial y las particularidades de la evaluación de estos sistemas.

    • English

      In the age of large language models, artificial intelligence’s goal has evolved to assist humans in unprecedented ways. As LLMs integrate into society, the need for comprehensive evaluations increases. These systems’ real-world acceptance depends on their knowledge, reasoning, and argumentation abilities. However, inconsistent standards across domains complicate evaluations, making it hard to compare models and understand their pros and cons. Our study focuses on illuminating the evaluation processes for these models. We examine recent research, tracking current trends to ensure evaluation methods match the field’s rapid progress requirements. We analyze key evaluation dimensions, aiming to deeply understand factors affecting models performance. A key aspect of our work is identifying and compiling major performance challenges and hazards in evaluation, an area not extensively explored yet. This approach is necessary for recognizing the potential and limitations of these AI systems in various domains of the evaluation.

  • Referencias bibliográficas
    • Aftan, S. and H. Shah. 2023. A survey on bert and its applications. In 2023 20th Learning and Technology Conference (L&T), pages 161–166....
    • Aiyappa, R., J. An, H. Kwak, and Y.-Y. Ahn. 2023. Can we trust the evaluation on chatgpt? arXiv preprint arXiv:2303.12767.
    • Awasthi, I., K. Gupta, P. S. Bhogal, S. S. Anand, and P. K. Soni. 2021. Natural language processing (nlp) based text summarization-a survey....
    • Baltrusaitis, T., C. Ahuja, and L.-P. Morency. 2018. Multimodal machine learning: A survey and taxonomy. IEEE transactions on pattern analysis...
    • Bang, Y., S. Cahyawijaya, N. Lee, W. Dai, D. Su, B. Wilie, H. Lovenia, Z. Ji, T. Yu, W. Chung, et al. 2023. A multitask, multilingual, multimodal...
    • Baradaran, R., R. Ghiasi, and H. Amirkhani. 2022. A survey on machine reading comprehension systems. Natural Language Engineering, 28(6):683–732.
    • Baroiu, A.-C. and S. Trausan-Matu. 2023. How capable are state-of-the-art language models to cope with sarcasm? In 2023 24th International...
    • Bates, M. 1995. Models of natural language understanding. Proceedings of the National Academy of Sciences, 92(22):9977–9982.
    • Berglund, L., M. Tong, M. Kaufmann, M. Balesni, A. C. Stickland, T. Korbak, and O. Evans. 2023. The reversal curse: Llms trained on.a is b”fail...
    • Bommasani, R., D. A. Hudson, E. Adeli, R. Altman, S. Arora, S. von Arx, M. S. Bernstein, J. Bohg, A. Bosselut, E. Brunskill, et al. 2021....
    • Bouziane, A., D. Bouchiha, N. Doumi, and M. Malki. 2015. Question answering systems: survey and trends. Procedia Computer Science, 73:366–375.
    • Bradbury, J., R. Frostig, P. Hawkins, M. J. Johnson, C. Leary, D. Maclaurin, G. Necula, A. Paszke, J. VanderPlas, S. Wanderman-Milne, et al....
    • Brown, T., B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sastry, A. Askell, et al. 2020. Language...
    • Bubeck, S., V. Chandrasekaran, R. Eldan, J. Gehrke, E. Horvitz, E. Kamar, P. Lee, Y. T. Lee, Y. Li, S. Lundberg, et al. 2023. Sparks of artificial...
    • Buchanan, B. G. and E. H. Shortliffe. 1984. Rule based expert systems: the mycin experiments of the stanford heuristic programming project...
    • Chang, Y., X. Wang, J. Wang, Y. Wu, K. Zhu, H. Chen, L. Yang, X. Yi, C. Wang, Y. Wang, et al. 2023. A survey on evaluation of large language...
    • Chen, M., J. Tworek, H. Jun, Q. Yuan, H. P. d. O. Pinto, J. Kaplan, H. Edwards, Y. Burda, N. Joseph, G. Brockman, et al. 2021. Evaluating...
    • Chen, Y., A. Arunasalam, and Z. B. Celik. 2023. Can large language models provide security & privacy advice? measuring the ability of...
    • Chetnani, Y. P. 2023. Evaluating the Impact of Model Size on Toxicity and Stereotyping in Generative LLM. Ph.D. thesis, State University of...
    • Chiang, C.-H. and H.-y. Lee. 2023. Can large language models be an alternative to human evaluations? arXiv preprint arXiv:2305.01937.
    • Chollet, F. 2019. On the measure of intelligence. arXiv preprint arXiv:1911.01547.
    • Chowdhery, A., S. Narang, J. Devlin, M. Bosma, G. Mishra, A. Roberts, P. Barham, H. W. Chung, C. Sutton, S. Gehrmann, et al. 2022. Palm: Scaling...
    • Clark, E., S. Rijhwani, S. Gehrmann, J. Maynez, R. Aharoni, V. Nikolaev, T. Sellam, A. Siddhant, D. Das, and A. P. Parikh. 2023. Seahorse:...
    • Cobbe, K., V. Kosaraju, M. Bavarian, M. Chen, H. Jun, L. Kaiser, M. Plappert, J. Tworek, J. Hilton, R. Nakano, et al. 2021. Training verifiers...
    • Costa-jussà, M. R., J. Cross, O. Çelebi, M. Elbayad, K. Heafield, K. Heffernan, E. Kalbassi, J. Lam, D. Licht, J. Maillard, et al. 2022. No...
    • Costantini, S. 2002. Meta-reasoning: A survey. In Computational Logic: Logic Programming and Beyond: Essays in Honour of Robert A. Kowalski...
    • Creswell, A., M. Shanahan, and I. Higgins. 2022. Selection-inference: Exploiting large language models for interpretable logical reasoning....
    • de Wynter, A., X. Wang, A. Sokolov, Q. Gu, and S.-Q. Chen. 2023. An evaluation on large language model outputs: Discourse and memorization....
    • Demarco, F., J. M. O. de Zarate, and E. Feuerstein. 2023. Measuring ideological spectrum through nlp. In Proceedings of the Seventh Workshop...
    • Demetriadis, S. and Y. Dimitriadis. 2023. Conversational agents and language models that learn from human dialogues to support design thinking....
    • Deng, J. and Y. Lin. 2022. The benefits and challenges of chatgpt: An overview. Frontiers in Computing and Intelligent Systems, 2(2):81–83.
    • Devlin, J., M.-W. Chang, K. Lee, and K. Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding....
    • Dong, G., J. Zhao, T. Hui, D. Guo, W. Wang, B. Feng, Y. Qiu, Z. Gongque, K. He, Z. Wang, et al. 2023. Revisit input perturbation problems...
    • Erdem, E., M. Kuyu, S. Yagcioglu, A. Frank, L. Parcalabescu, B. Plank, A. Babii, O. Turuta, A. Erdem, I. Calixto, et al. 2022. Neural natural...
    • Floridi, L. and M. Chiriatti. 2020. Gpt-3: Its nature, scope, limits, and consequences. Minds and Machines, 30:681–694.
    • Frieder, S., L. Pinchetti, R.-R. Griffiths, T. Salvatori, T. Lukasiewicz, P. C. Petersen, A. Chevalier, and J. Berner. 2023. Mathematical...
    • Fu, Y., H. Peng, and T. Khot. 2022. How does gpt obtain its ability? tracing emergent abilities of language models to their sources. Yao Fu’s...
    • Gamallo, P., J. R. P. Campos, and I. Alegria. 2017. A perplexity-based method for similar languages discrimination. In Proceedings of the...
    • Gao, J. and C.-Y. Lin. 2004. Introduction to the special issue on statistical language modeling. ACM Transactions on Asian Language Information...
    • Garg, T., S. Masud, T. Suresh, and T. Chakraborty. 2023. Handling bias in toxic speech detection: A survey. ACM Computing Surveys, 55(13s):1–32.
    • Ghazal, A., T. Rabl, M. Hu, F. Raab, M. Poess, A. Crolotte, and H.-A. Jacobsen. 2013. Bigbench: Towards an industry standard benchmark for...
    • Hadi, M. U., R. Qureshi, A. Shah, M. Irfan, A. Zafar, M. Shaikh, N. Akhtar, J.Wu, and S. Mirjalili. 2023a. A survey on large language models:...
    • Hadi, M. U., R. Qureshi, A. Shah, M. Irfan, A. Zafar, M. B. Shaikh, N. Akhtar, J. Wu, S. Mirjalili, et al. 2023b. Large language models: a...
    • Head, C. B., P. Jasper, M. McConnachie, L. Raftree, and G. Higdon. 2023. Large language model applications for evaluation: Opportunities and...
    • Hendrycks, D., C. Burns, S. Basart, A. Zou, M. Mazeika, D. Song, and J. Steinhardt. 2020. Measuring massive multitask language understanding....
    • Hoffmann, J., S. Borgeaud, A. Mensch, E. Buchatskaya, T. Cai, E. Rutherford, D. d. L. Casas, L. A. Hendricks, J. Welbl, A. Clark, et al. 2022....
    • Huang, D., Q. Bu, J. Zhang, X. Xie, J. Chen, and H. Cui. 2023. Bias assessment and mitigation in llm-based code generation. arXiv preprint...
    • Jain, N., K. Saifullah, Y. Wen, J. Kirchenbauer, M. Shu, A. Saha, M. Goldblum, J. Geiping, and T. Goldstein. 2023. Bring your own data! self-supervised...
    • Ji, Z., N. Lee, R. Frieske, T. Yu, D. Su, Y. Xu, E. Ishii, Y. J. Bang, A. Madotto, and P. Fung. 2023. Survey of hallucination in natural language...
    • Jin, Z., J. Liu, Z. Lyu, S. Poff, M. Sachan, R. Mihalcea, M. Diab, and B. Schölkopf. 2023. Can large language models infer causation from...
    • Kaplan, J., S. McCandlish, T. Henighan, T. B. Brown, B. Chess, R. Child, S. Gray, A. Radford, J. Wu, and D. Amodei. 2020. Scaling laws for...
    • Kasneci, E., K. Seßler, S. K¨uchemann, M. Bannert, D. Dementieva, F. Fischer, U. Gasser, G. Groh, S. G¨unnemann, E. Hüllermeier, et al. 2023....
    • Kejriwal, M., H. Santos, K. Shen, A. M. Mulvehill, and D. L. McGuinness. 2023. Context-rich evaluation of machine common sense. In International...
    • Khalfa, J. 1994. What is intelligence? Cambridge University Press.
    • Khowaja, S. A., P. Khuwaja, and K. Dev. 2023. Chatgpt needs spade (sustainability, privacy, digital divide, and ethics) evaluation: A review....
    • Koh, J. Y., R. Salakhutdinov, and D. Fried. 2023. Grounding language models to images for multimodal generation. arXiv preprint arXiv:2301.13823.
    • Korb, K. B. and A. E. Nicholson. 2010. Bayesian artificial intelligence. CRC press.
    • Kotek, H., R. Dockum, and D. Sun. 2023. Gender bias and stereotypes in large language models. In Proceedings of The ACM Collective Intelligence...
    • Lacave, C. and F. J. Díez. 2002. A review of explanation methods for bayesian networks. The Knowledge Engineering Review, 17(2):107–127.
    • Lai, V. D., N. T. Ngo, A. P. B. Veyseh, H. Man, F. Dernoncourt, T. Bui, and T. H. Nguyen. 2023. Chatgpt beyond english: Towards a comprehensive...
    • Lazarski, E., M. Al-Khassaweneh, and C. Howard. 2021. Using nlp for fact checking: A survey. Designs, 5(3):42.
    • Lehman, J., J. Gordon, S. Jain, K. Ndousse, C. Yeh, and K. O. Stanley. 2023. Evolution through large models. In Handbook of Evolutionary Machine...
    • Li, J., T. Tang, W. X. Zhao, J.-Y. Nie, and J.-R. Wen. 2022. Pretrained language models for text generation: A survey. arXiv preprint arXiv:2201.05273.
    • Li, Y., M. Du, R. Song, X. Wang, and Y. Wang. 2023. A survey on fairness in large language models. arXiv preprint arXiv:2308.10149.
    • Liang, P., R. Bommasani, T. Lee, D. Tsipras, D. Soylu, M. Yasunaga, Y. Zhang, D. Narayanan, Y. Wu, A. Kumar, et al. 2022. Holistic evaluation...
    • Lin, S., J. Hilton, and O. Evans. 2021. Truthfulqa: Measuring how models mimic human falsehoods. arXiv preprint arXiv:2109.07958.
    • Liu, F., E. Bugliarello, E. M. Ponti, S. Reddy, N. Collier, and D. Elliott. 2021. Visually grounded reasoning across languages and cultures....
    • Ma, J.-Y., J.-C. Gu, Z.-H. Ling, Q. Liu, and C. Liu. 2023. Untying the reversal curse via bidirectional language model editing. arXiv preprint...
    • Mahany, A., H. Khaled, N. S. Elmitwally, N. Aljohani, and S. Ghoniemy. 2022. Negation and speculation in nlp: A survey, corpora, methods,...
    • McDonald, D. D. 2010. Natural language generation. Handbook of natural language processing, 2:121–144.
    • Min, B., H. Ross, E. Sulem, A. P. B. Veyseh, T. H. Nguyen, O. Sainz, E. Agirre, I. Heintz, and D. Roth. 2023. Recent advances in natural language...
    • Motger, Q., X. Franch, and J. Marco. 2022. Software-based dialogue systems: survey, taxonomy, and challenges. ACM Computing Surveys, 55(5):1–42.
    • Nijkamp, E., H. Hayashi, C. Xiong, S. Savarese, and Y. Zhou. 2023. Codegen2: Lessons for training llms on programming and natural languages....
    • Novikova, J., O. Duˇsek, A. C. Curry, and V. Rieser. 2017. Why we need new evaluation metrics for nlg. arXiv preprint arXiv:1707.06875.
    • OpenAI, R. 2023. Gpt-4 technical report. Arxiv 2303.08774. View in Article, 2.
    • Orrù, G., A. Piarulli, C. Conversano, and A. Gemignani. 2023. Human-like problem-solving abilities in large language models using chatgpt....
    • Oshikawa, R., J. Qian, and W. Y.Wang. 2018. A survey on natural language processing for fake news detection. arXiv preprint arXiv:1811.00770.
    • Ouyang, L., J. Wu, X. Jiang, D. Almeida, C. Wainwright, P. Mishkin, C. Zhang, S. Agarwal, K. Slama, A. Ray, et al. 2022. Training language...
    • Peng, Z., Z. Wang, and D. Deng. 2023. Nearduplicate sequence search at scale for large language model memorization evaluation. Proceedings...
    • Perez, E., S. Huang, F. Song, T. Cai, R. Ring, J. Aslanides, A. Glaese, N. McAleese, and G. Irving. 2022. Red teaming language models with...
    • Puchert, P., P. Poonam, C. van Onzenoodt, and T. Ropinski. 2023. Llmmaps–a visual metaphor for stratified evaluation of large language models....
    • Qin, C., A. Zhang, Z. Zhang, J. Chen, M. Yasunaga, and D. Yang. 2023. Is chatgpt a general-purpose natural language processing task solver?...
    • Reiter, E. 2018. A structured review of the validity of bleu. Computational Linguistics, 44(3):393–401.
    • Rillig, M. C., M. ˚Agerstrand, M. Bi, K. A. Gould, and U. Sauerland. 2023. Risks and benefits of large language models for the environment....
    • Ruder, S., J. H. Clark, A. Gutkin, M. Kale, M. Ma, M. Nicosia, S. Rijhwani, P. Riley, J.-M. A. Sarr, X. Wang, et al. 2023. Xtreme-up: A user-centric...
    • Saha, T., D. Ganguly, S. Saha, and P. Mitra. 2023. Workshop on large language models’ interpretability and trustworthiness (llmit). In Proceedings...
    • Sainz, O., J. A. Campos, I. García-Ferrero, J. Etxaniz, O. L. de Lacalle, and E. Agirre. 2023. Nlp evaluation in trouble: On the need to measure...
    • Sakaguchi, K., R. L. Bras, C. Bhagavatula, and Y. Choi. 2021. Winogrande: An adversarial winograd schema challenge at scale. Communications...
    • Salloum, S. A., R. Khan, and K. Shaalan. 2020. A survey of semantic analysis approaches. In Proceedings of the International Conference on...
    • Shanahan, M. 2022. Talking about large language models. arXiv preprint arXiv:2212.03551.
    • Shin, J. and J. Nam. 2021. A survey of automatic code generation from natural language. Journal of Information Processing Systems, 17(3):537–555.
    • Song, G., Y. Ye, X. Du, X. Huang, and S. Bie. 2014. Short text classification: a survey. Journal of multimedia, 9(5).
    • Srivastava, A., A. Rastogi, A. Rao, A. A. M. Shoeb, A. Abid, A. Fisch, A. R. Brown, A. Santoro, A. Gupta, A. Garriga-Alonso, et al. 2022....
    • Storks, S., Q. Gao, and J. Y. Chai. 2019. Recent advances in natural language inference: A survey of benchmarks, resources, and approaches....
    • Sun, J., S. Wang, J. Zhang, and C. Zong. 2020. Distill and replay for continual language learning. In Proceedings of the 28th international...
    • Tang, R., Y.-N. Chuang, and X. Hu. 2023. The science of detecting llm-generated texts. arXiv preprint arXiv:2303.07205.
    • Tedeschi, S., J. Bos, T. Declerck, J. Hajic, D. Hershcovich, E. H. Hovy, A. Koller, S. Krek, S. Schockaert, R. Sennrich, et al. 2023. What’s...
    • Touvron, H., T. Lavril, G. Izacard, X. Martinet, M.-A. Lachaux, T. Lacroix, B. Rozière, N. Goyal, E. Hambro, F. Azhar, et al. 2023. Llama:...
    • Vaswani, A., N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin. 2017. Attention is all you need. Advances...
    • Wang, A., Y. Pruksachatkun, N. Nangia, A. Singh, J. Michael, F. Hill, O. Levy, and S. Bowman. 2019. Superglue: A stickier benchmark for general-purpose...
    • Wang, A., A. Singh, J. Michael, F. Hill, O. Levy, and S. R. Bowman. 2018. Glue: A multi-task benchmark and analysis platform for natural language...
    • Wang, X., J. Wei, D. Schuurmans, Q. Le, E. Chi, S. Narang, A. Chowdhery, and D. Zhou. 2022. Self-consistency improves chain of thought reasoning...
    • Wang, Y., Y. Wang, J. Liu, and Z. Liu. 2020. A comprehensive survey of grammar error correction. arXiv preprint arXiv:2005.06600.
    • Wei, J., Y. Tay, R. Bommasani, C. Raffel, B. Zoph, S. Borgeaud, D. Yogatama, M. Bosma, D. Zhou, D. Metzler, et al. 2022. Emergent abilities...
    • Xu, F. F., U. Alon, G. Neubig, and V. J. Hellendoorn. 2022. A systematic evaluation of large language models of code. In Proceedings of the...
    • Xu, P., W. Shao, K. Zhang, P. Gao, S. Liu, M. Lei, F. Meng, S. Huang, Y. Qiao, and P. Luo. 2023a. Lvlm-ehub: A comprehensive evaluation benchmark...
    • Xu, X., K. Kong, N. Liu, L. Cui, D. Wang, J. Zhang, and M. Kankanhalli. 2023b. An llm can fool itself: A prompt-based adversarial attack....
    • Zellers, R., A. Holtzman, Y. Bisk, A. Farhadi, and Y. Choi. 2019. Hellaswag: Can a machine really finish your sentence? arXiv preprint arXiv:1905.07830.
    • Zhai, Y., S. Tong, X. Li, M. Cai, Q. Qu, Y. J. Lee, and Y. Ma. 2023. Investigating the catastrophic forgetting in multimodal large language...
    • Zhang, L., S. Wang, and B. Liu. 2018. Deep learning for sentiment analysis: A survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge...
    • Zhang, R., J. Han, A. Zhou, X. Hu, S. Yan, P. Lu, H. Li, P. Gao, and Y. Qiao. 2023. Llama-adapter: Efficient fine-tuning of language models...
    • Zhao, W. X., K. Zhou, J. Li, T. Tang, X. Wang, Y. Hou, Y. Min, B. Zhang, J. Zhang, Z. Dong, et al. 2023. A survey of large language models....
    • Zhong, W., R. Cui, Y. Guo, Y. Liang, S. Lu, Y.Wang, A. Saied, W. Chen, and N. Duan. 2023. Agieval: A human-centric benchmark for evaluating...
    • Zhu, K., J. Wang, J. Zhou, Z. Wang, H. Chen, Y. Wang, L. Yang, W. Ye, N. Z. Gong, Y. Zhang, et al. 2023. Promptbench: Towards evaluating the...
Fundación Dialnet

Mi Documat

Opciones de artículo

Opciones de compartir

Opciones de entorno