Paper page - PlantMarkerBench: A Multi-Species Benchmark for Evidence-Grounded Plant Marker Reasoning
… We benchmark diverse open-weight and closed-source language models across species and prompting strategies. …
… We benchmark diverse open-weight and closed-source language models across species and prompting strategies. …
… The versions of the libraries I used are: trl==0.18.2 peft==0.15.2 transformers==4.52.4 deepspeed==0.17.1 LoRA Config lora config = LoraConfig r=training config "rank" , lora alpha=training config "alpha" , target modules= "q proj", "k proj", "v proj", "o proj", "gate proj", "up proj", "down proj",… …
… I tried prompting with scraping strategies but in the end hardly any meaningful results were found. …