List your company

Single-hop Tool Calling on WHEN2TOOL single-hop 1.0 (test)

94.3AccuracyPROBE&PREFILL
23.5841.9460.378.66May 10, 2026

Evaluation results

MethodLinks
PROBE&PREFILLBackbone=Qwen3-14B, tau=0.12026.0594.32,128
Prompt-onlyBackbone=Qwen3-32B, Setting=Default2026.0594.12,404
PROBE&PREFILLBackbone=Qwen3-14B, tau=0.32026.0594.11,509
PROBE&PREFILLBackbone=Qwen3-32B, tau=0.12026.05941,996
Prompt-onlyBackbone=Qwen3-14B, Setting=Force2026.0593.72,421
Prompt-onlyBackbone=Qwen3-14B, Setting=Default2026.0593.72,211
Reason-then-ActBackbone=Qwen3-32B, Setting=Force2026.0593.72,150
Prompt-onlyBackbone=Qwen3-14B, Setting=Necessary2026.0593.52,175
Prompt-onlyBackbone=Qwen3-32B, Setting=Necessary2026.0593.52,394
Reason-then-ActBackbone=Qwen3-14B, Setting=Force2026.0593.42,172
Prompt-onlyBackbone=Qwen3-32B, Setting=Sparse2026.0593.31,912
Reason-then-ActBackbone=Qwen3-32B, Setting=Necessary2026.0593.21,845
PROBE&PREFILLBackbone=Qwen3-32B, tau=0.32026.0593.21,493
Reason-then-ActBackbone=Qwen3-32B, Setting=Default2026.0592.91,823
Prompt-onlyBackbone=Qwen3-32B, Setting=Force2026.0592.82,559
Reason-then-ActBackbone=Qwen3-14B, Setting=Necessary2026.0592.41,634
PROBE&PREFILLBackbone=Qwen3-14B, tau=0.52026.0592.41,227
Prompt-onlyBackbone=Qwen3-4B, Setting=Force2026.05922,435
Reason-then-ActBackbone=Qwen3-14B, Setting=Default2026.05921,589
PROBE&PREFILLBackbone=Qwen3-4B, tau=0.12026.0591.12,216
Reason-then-ActBackbone=Qwen3-4B, Setting=Force2026.0590.91,651
PROBE&PREFILLBackbone=Qwen3-4B, tau=0.32026.0590.41,707
PROBE&PREFILLBackbone=Qwen3-32B, tau=0.52026.0590.11,155
Prompt-onlyBackbone=Qwen3-4B, Setting=Default2026.0589.22,118
PROBE&PREFILLBackbone=Llama-70B, tau=0.92026.0589.22,804
PROBE&PREFILLBackbone=Qwen3-1.7B, tau=0.32026.05892,507
PROBE&PREFILLBackbone=Qwen3-1.7B, tau=0.12026.0588.82,512
PROBE&PREFILLBackbone=Llama-70B, tau=0.32026.0588.62,902
PROBE&PREFILLBackbone=Llama-70B, tau=0.72026.0588.62,828
PROBE&PREFILLBackbone=Qwen3-4B, tau=0.52026.0588.51,309
PROBE&PREFILLBackbone=Llama-70B, tau=0.12026.0588.42,976
PROBE&PREFILLBackbone=Qwen3-1.7B, tau=0.52026.0588.32,128
PROBE&PREFILLBackbone=Llama-70B, tau=0.52026.0588.32,871
Prompt-onlyBackbone=Qwen3-1.7B, Setting=Default2026.0588.22,709
Reason-then-ActBackbone=Qwen3-32B, Setting=Sparse2026.0587.91,231
Prompt-onlyBackbone=Qwen3-1.7B, Setting=Necessary2026.0587.22,787
Prompt-onlyBackbone=Qwen3-1.7B, Setting=Force2026.0586.83,120
Reason-then-ActBackbone=Qwen3-1.7B, Setting=Force2026.0586.72,475
PROBE&PREFILLBackbone=Qwen3-14B, tau=0.72026.0585.8907
Prompt-onlyBackbone=Qwen3-4B, Setting=Necessary2026.0585.71,852
Reason-then-ActBackbone=Qwen3-14B, Setting=Sparse2026.0585.51,034
PROBE&PREFILLBackbone=Qwen3-4B, tau=0.72026.0584.81,026
Reason-then-ActBackbone=Qwen3-1.7B, Setting=Default2026.0584.71,923
Prompt-onlyBackbone=Llama-70B, Setting=Necessary2026.0584.43,988
Prompt-onlyBackbone=Llama-70B, Setting=Sparse2026.0584.23,431
Reason-then-ActBackbone=Qwen3-1.7B, Setting=Necessary2026.05841,871
Reason-then-ActBackbone=Qwen3-14B, Setting=No-tool2026.0583.5928
Reason-then-ActBackbone=Qwen3-4B, Setting=Default2026.0583.41,024
Reason-then-ActBackbone=Qwen3-4B, Setting=Necessary2026.0583.41,027
Prompt-onlyBackbone=Llama-70B, Setting=Default2026.0583.14,377
Prompt-onlyBackbone=Llama-70B, Setting=No-tool2026.0582.53,615
PROBE&PREFILLBackbone=Qwen3-32B, tau=0.72026.0582.3896
PROBE&PREFILLBackbone=Qwen3-1.7B, tau=0.72026.0581.61,415
Prompt-onlyBackbone=Llama-8B, Setting=Sparse2026.0581.42,868
Prompt-onlyBackbone=Llama-8B, Setting=Default2026.0579.53,708
Prompt-onlyBackbone=Llama-8B, Setting=Force2026.0578.74,081
Prompt-onlyBackbone=Llama-70B, Setting=Force2026.0578.35,302
Prompt-onlyBackbone=Llama-8B, Setting=Necessary2026.0577.73,548
Prompt-onlyBackbone=Llama-8B, Setting=No-tool2026.0577.63,391
Prompt-onlyBackbone=Qwen3-14B, Setting=Sparse2026.0577.31,020
PROBE&PREFILLBackbone=Qwen3-4B, tau=0.92026.0574.7657
Reason-then-ActBackbone=Qwen3-32B, Setting=No-tool2026.0574.3596
Reason-then-ActBackbone=Qwen3-1.7B, Setting=Sparse2026.0573.21,161
Reason-then-ActBackbone=Qwen3-4B, Setting=Sparse2026.0572.5535
PROBE&PREFILLBackbone=Qwen3-32B, tau=0.92026.0571.5604
Reason-then-ActBackbone=Qwen3-1.7B, Setting=No-tool2026.0571.11,117
Prompt-onlyBackbone=Qwen3-4B, Setting=Sparse2026.0570.7484
Prompt-onlyBackbone=Qwen3-1.7B, Setting=Sparse2026.0570.21,611
PROBE&PREFILLBackbone=Llama-8B, tau=0.52026.0569.72,381
PROBE&PREFILLBackbone=Llama-8B, tau=0.12026.0569.23,027
PROBE&PREFILLBackbone=Llama-8B, tau=0.32026.0568.92,770
Prompt-onlyBackbone=Qwen3-32B, Setting=No-tool2026.0566.8451
PROBE&PREFILLBackbone=Llama-8B, tau=0.72026.0566.52,146
PROBE&PREFILLBackbone=Qwen3-14B, tau=0.92026.0566347
Reason-then-ActBackbone=Qwen3-4B, Setting=No-tool2026.0564.5315
PROBE&PREFILLBackbone=Llama-8B, tau=0.92026.0561.71,753
Reason-then-ActBackbone=Llama-70B, Setting=No-tool2026.0554.80
Reason-then-ActBackbone=Llama-70B, Setting=Sparse2026.0554.10
Prompt-onlyBackbone=Qwen3-14B, Setting=No-tool2026.0551.285
Prompt-onlyBackbone=Qwen3-4B, Setting=No-tool2026.0550.8121
Reason-then-ActBackbone=Llama-70B, Setting=Default2026.0547.90
PROBE&PREFILLBackbone=Qwen3-1.7B, tau=0.92026.0547.9293
Reason-then-ActBackbone=Llama-70B, Setting=Necessary2026.0547.60
Reason-then-ActBackbone=Llama-70B, Setting=Force2026.0542.40
Reason-then-ActBackbone=Llama-8B, Setting=Sparse2026.0536.30
Reason-then-ActBackbone=Llama-8B, Setting=No-tool2026.0534.56
Reason-then-ActBackbone=Llama-8B, Setting=Necessary2026.053211
Reason-then-ActBackbone=Llama-8B, Setting=Default2026.0531.22
Reason-then-ActBackbone=Llama-8B, Setting=Force2026.0529.39
Prompt-onlyBackbone=Qwen3-1.7B, Setting=No-tool2026.0526.3293