arxiv:2606.00408
XuQixin
Racktic
AI & ML interests
NLP, mutimodel
Recent Activity
authored a paper 8 days ago
Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism upvoted a paper 10 days ago
Harness-1: Reinforcement Learning for Search Agents with State-Externalizing HarnessesOrganizations
models 10
Racktic/DSGym-models
Updated
Racktic/scientist_7b_all_4o_full_toolset_step10
Updated
Racktic/stage2_qwen_distill_sft
8B • Updated • 2
Racktic/stage2_sft554
8B • Updated • 7
Racktic/stage2_bestrl
8B • Updated • 7
Racktic/stage1_11k_2epoch
8B • Updated • 1
Racktic/stage1_11k_stage2_11k_mix_step400
8B • Updated • 1
Racktic/doc_val
Updated
Racktic/mix_sft_step800
8B • Updated • 2
Racktic/qwen_dedup_top8_dpo_old_math_new_math_syn_olympiads
Updated
datasets 26
Racktic/DeepSearch-Attn
Updated • 2.54k
Racktic/dsgym_backup
Updated • 237
Racktic/doc_val_outputs
Preview • Updated • 112
Racktic/ablation_nosyndata_logs
Preview • Updated • 160
Racktic/dude_origin_train
Viewer • Updated • 6k • 17
Racktic/ablation_nosyndata
Preview • Updated • 142
Racktic/zeroRL
Preview • Updated • 139
Racktic/sft554_rl
Preview • Updated • 149
Racktic/novcentric_rl
Preview • Updated • 103
Racktic/abl_11ks1
Preview • Updated • 125