Tool-Calling LLMs — Page 949

Qwen3-1.7B-Base-dapo_filter-grpo-useKL_True-KLlossCoef1e-3

Qwen3-8B-tacq-4bit-calibration-Chinese-128samples

KG-R1-CWQ-hit1-no-turn-advantage

V3ra-Insync-AI-v3-merged

qwen2.5_sft_merged_dk_it

akeel-4B-lora

llama-3_1-8b-simnpo-gentle-bm25-10b

qwen3-8b-simnpo-gentle-bm25-10b

KG-R1-CWQ-no-retrieval-reward

Llama-3.2-3B-Instruct_grpo_ppl_adv_rollout_8_20260429_004543_step580

g1_top8_diverse_100000_32b__Qwen3-32B

g1_top8_85k_gptlong_swegym_32b_step3600__Qwen3-32B

gptlong_continue_top8diverse100k_step2700__Qwen3-32B

tezos100k_continue_gptlongtezos_step900__Qwen3-32B

vid_score_qwen3_8b_lora16_hifps_doverref_merged_step3040

vid_score_qwen3_8b_lora16_hires_doverref_merged_step3040

Qwen3-1.7B-Base-dapo_filter-grpo-noKL

Qwen_plus2_shot7_sft_fold0

llama-3_1-8b-undial-baseline-target-100

affine-T55-5EWd7djizaL8bq78dN8PqsMm4UVvdGrfBsToKroHBzgFs2QP

qwen3-8b-simnpo-gentle-igm-10b

DialFactSum-Base-8B

triage_mistral_finetuned

math_m32-4b-9e032637-not_easy_1e-4_800

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_bio_sensitive20pct_nogap-maxsteps150

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_material_middle20_nogap-maxsteps150-h

turkish-ecommerce-aspect-summarizer

llama3_2_3b_instruct_MATH_lr5e-5

qwen2.5_math_1.5b_grpo_ppl_adv_step580

Meta-Llama-3-8B-SFT-safe

filing-sense-grpo-qwen2.5-3b

math_skywork-v2-qwen3-4b-easy_1e-4_200

llama-3.1-8B-gsm8k-rsn-tuned-lr5e-5

evolai-qwen2.5-1.5b-sn47-v2

fake_english_advshape_policyshape_qwen3-1.7b-base

1B-Instruct-Tulu-full

Affine-5FbLST7rfr8sugrJHkJFJYLxkHhvVPY1qbnWPuDUrYArjA6y

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_chem_middle20_nogap-maxsteps150

affine-5F4JyqstSdvMfZcRuFvyAGPer25Cu1PmNd3snnHfaA7gxguZ

llama-3_1-8b-simnpo-gentle-baseline-target-100

zay-qwen15-text2cypher-lotob-v1

Qwen3-1.7B-Base-dapo_filter-prm-eta100-Advorm-stepsplit-none